Pedicled, vascularized occipital bone graft to supplement atlantoaxial arthrodesis for the treatment of pseudoarthrosis

Journal of Clinical Neuroscience xxx (xxxx) xxx

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn

Pedicled, vascularized occipital bone graft to supplement atlantoaxial arthrodesis for the treatment of pseudoarthrosis Edward M. Reece a, Aditya Vedantam b, Sungho Lee b, Mohin Bhadkamkar a, Matthew Kaufman a, Michael A. Bohl c, Steve W. Chang c, Randall W. Porter c, Nicholas Theodore d, U. Kumar Kakarla c, Alexander E. Ropper b,⇑ a

Division of Plastic Surgery, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, TX, USA Department of Neurosurgery, Baylor College of Medicine, Houston, TX, USA Department of Neurosurgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, AZ, USA d Department of Neurosurgery, Johns Hopkins Hospital, Baltimore, MD, USA b c

a r t i c l e

i n f o

Article history: Received 19 November 2018 Accepted 12 April 2019 Available online xxxx Keywords: Atlantoaxial arthrodesis Pseudoarthrosis Revision spine surgery Transarticular fixation Vascularized bone grafts Suboccipital craniectomy

a b s t r a c t Atlantoaxial pseudoarthrosis is a challenging postoperative complication. The use of a local, vascularized bone graft, without free tissue transfer, to support a revision atlantoaxial fusion has not been previously described. We report the first surgical patient who received a semispinalis capitis muscle pedicled, occipital bone graft for supplementation of a revision atlantoaxial arthrodesis. A 72-year-old female had a failed atlantoaxial fusion and developed neck pain from continued instability and fractured hardware. The fixation and fusion were revised and supplemented with a novel, pedicled occipital bone graft. A craniectomy was performed in the occipital bone while still attached to the semispinalis capitis muscle to provide graft vascularity. This graft was rotated inferiorly from the skull base to the C1 arch and C2 spinous process in order to supplement a revision atlantoaxial arthrodesis. The patient had excellent clinical recovery over 18-month clinical follow up. The bone graft harvesting and rotation were performed safely and without complication. The 6-month postoperative CT scan showed partial fusion into the graft. This novel surgical technique leverages the advantages of vascularized structural autograft without adding extensive time or morbidity to the procedure as observed in free-tissue transfers. It is a safe and useful salvage technique to supplement revision atlantoaxial fusion surgeries. Ó 2019 Elsevier Ltd. All rights reserved.

1. Introduction Successful arthrodesis is critical for achieving favorable outcomes in patients undergoing craniocervical and atlantoaxial stabilization. For many patients, non-vascularized bone grafts (N-VBGs) are routinely used. However, for those with comorbidities impairing bony fusion or those with a history of pseudoarthrosis, N-VBGs may be insufficient for achieving successful arthrodesis. An alternative to free-transfer vascularized bone grafts (VBG) is pedicled VBG that is composed of a locally harvested piece of bone, rotated into the fusion bed while maintaining its own blood supply

through muscle and periosteal vessels. Wilden et al. [1] and Lewis et al. [2] reported success with pedicled rib grafts in thoracic spine reconstruction. They showed that harvest of a vascularized rib graft and mobilization of a vascular pedicle avoided a vessel anastomosis and added less than 1 h of operative time. Pedicled VBG combines benefits of VBGs while avoiding the challenges and morbidity of free-tissue transfer and may be an attractive treatment option for patients with, or at increased risk of, pseudarthrosis. We describe the successful utilization of a novel, pedicled occipital VBG to supplement atlantoaxial fusion.

2. Technical report Abbreviations: N-VBG, non-vascularized bone graft; O-VBG, occipitalvascularized bone graft; VBG, vascularized bone graft. ⇑ Corresponding author at: Department of Neurosurgery, Baylor College of Medicine, 7200 Cambridge St. Suite 9A, Houston, TX, USA. E-mail address: [email protected] (A.E. Ropper).

The patient was a 72-year-old female, smoker with a history of axial neck pain, cervical radiculopathy and arm weakness. She originally presented with gradually worsening mechanical neck pain following a motor vehicle collision 6 months prior. Imaging

https://doi.org/10.1016/j.jocn.2019.04.014 0967-5868/Ó 2019 Elsevier Ltd. All rights reserved.

Please cite this article as: E. M. Reece, A. Vedantam, S. Lee et al., Pedicled, vascularized occipital bone graft to supplement atlantoaxial arthrodesis for the treatment of pseudoarthrosis, Journal of Clinical Neuroscience, https://doi.org/10.1016/j.jocn.2019.04.014

2

E.M. Reece et al. / Journal of Clinical Neuroscience xxx (xxxx) xxx

Fig. 1. Lateral cervical X-Rays performed prior to the index surgery, demonstrating an increased atlanto-dental interval (ADI) of 6 mm and instability at C1-2. Neutral (A), flexion (B), and extension (C) views.

revealed C1-2 instability with subaxial stenosis and instability (Fig. 1). An index surgery was performed to treat atlantoaxial instability and cervical stenosis with a C1-6 posterior fixation and arthrodesis and C4-6 decompression She recovered well from this surgery until 3 years post-operatively where she developed subacute worsening of mechanical neck pain. Imagining demonstrated

a C1-2 pseudoarthrosis with listhesis and bilateral C1 screw fractures. Although there appeared to be solid bony fusion from C2 to C6, there was no apparent fusion from C1 to C2 (Fig. 2). The patient’s bone density scan (DEXA) showed no osteopenia or osteoporosis. Revision surgery was recommended to treat the pseudoarthrosis and instrumentation failure. In addition, a

Fig. 2. Pre-operative imaging. Lateral cervical X-ray (A), axial CT (B), and reformatted right (C) and left (D) sagittal CT images demonstrate fractured C1 lateral mass screws and pseudoarthrosis of the atlantoaxial joint.

Please cite this article as: E. M. Reece, A. Vedantam, S. Lee et al., Pedicled, vascularized occipital bone graft to supplement atlantoaxial arthrodesis for the treatment of pseudoarthrosis, Journal of Clinical Neuroscience, https://doi.org/10.1016/j.jocn.2019.04.014

E.M. Reece et al. / Journal of Clinical Neuroscience xxx (xxxx) xxx

3

Fig. 3. Illustration showing harvest of vascularized occipital bone graft and rotation of the graft on its vascular pedicle into the space between C1 and C2. (Reprinted with permission; Baylor College of Medicine).

Fig. 4. Intraoperative photograph of the O-VBG in place with the revised fixation. The occipital bone graft (*) is visible beneath the cross-link. The arch of C1 is covered by the O-VBG in this picture. The semispinalis capitis muscle pedicle (^) is rotated from lateral to medial to supply the bone graft.

vascularized occipital bone graft would be harvested and used to supplement the revision C1-2 arthrodesis. Since her smoking was a major risk factor contributing to the original pseudoarthrosis, we required smoking and nicotine cessation prior to the revision surgery. Surgery was performed with intraoperative neuro-navigation following intraoperative CT. The patient was positioned prone in a Mayfield head holder (Integra LifeSciences; Plainsboro, NJ.) with the neck in neutral position. After reopening the prior incision, dissection was completed around the bilateral constructs. The old rods were removed and the C1 screw heads and proximal screw fragments were removed. In addition, the prior C2 pars screws and C3 lateral mass screws were removed. A power drill bit was navigated through C2, across the C1-2 joint and into the caudal portion of the C1 lateral masses. Following tapping, transarticular screws were placed to achieve atlantoaxial fixation. Additional soft tissue was dissected away and the C1-2 joint was decorticated and prepared for arthrodesis. Soft tissue dissection was extended superior to the C1 posterior arch and carried ventrally along the median raphe, taking care to preserve the posterior cervical musculature. Subperiosteal exposure of the median nuchal line of the occipital bone was performed. The intermuscular plane between the trapezius and semispinalis capitis muscle was then identified and dissected. A piece of occipital bone; bordered superiorly by the superior nuchal line, inferiorly by the foramen magnum, medially by the median nuchal line, and laterally by the limit of surgical exposure, was marked (Fig. 3, Reprinted with permission from Baylor College of Medicine). This segment of occipital bone was then further defined by subpe-

Please cite this article as: E. M. Reece, A. Vedantam, S. Lee et al., Pedicled, vascularized occipital bone graft to supplement atlantoaxial arthrodesis for the treatment of pseudoarthrosis, Journal of Clinical Neuroscience, https://doi.org/10.1016/j.jocn.2019.04.014

4

E.M. Reece et al. / Journal of Clinical Neuroscience xxx (xxxx) xxx

Fig. 5. Post-operative imaging. Axial CT (A), 3D reconstruction (B), and re-formatted right (C) and left (D) sagittal CT images demonstrate appropriate trajectory of the screws across the C1-2 lateral mass articulation bilaterally. Vascularized bone graft is visualized just behind the posterior arch of C1 in reformatted mid-sagittal CT (E; arrow) and 3D reconstruction images (B; arrow).

riosteal dissection around the borders only, taking care not to disrupt the central periosteum or semispinalis capitis muscle attachments. Burr holes were placed approximately 2 cm above the foramen magnum on the left side. A craniotomy was performed by connecting the burr holes with a craniotome and then releasing the lateral and medial sides of the bone flap, while protecting the central muscle pedicle. The inner cortex of the bone flap was gently decorticated, and bleeding was seen from the medullary bone, indicating residual blood supply from the pedicle. The flap was then rotated from the occiput over the C1 lamina and abutting and C2 while still attached to the rectus capitis posterior minor muscle. Prior to this rotation, the C1 and C2 lamina were decorticated. New titanium rods were placed in the screws from C2 to C6 and fashioned intentionally long at the rostral end to extend to the level of the C1 lamina. The area around the graft was packed with allograft. In order to hold the VBG in place and compress it against the C1 arch and the C2 spinous process, a crossconnector was placed above the heads of the transarticular screws (Fig. 4). The incision was then closed in a standard, layered fashion, after placement of a subfascial drain. Estimated blood loss was 100 cc and surgical time was 240 min. The patient had significant improvement in her neck pain on postoperative day 1 and was discharged on postoperative day 6 with a rigid cervical collar. A rigid cervical collar and bone stimulator were maintained for 6 months following surgery. A postoperative CT scan showed appropriate position of the O-VBG, however, it was separated from the dorsal, decorticated C1 arch by a small distance. The CT was repeated after three and six months, and there appeared to be some fibrous union between the O-VBG and the C2 spinous process (Fig. 5). There was no clear bony fusion yet with the C1 arch, but the graft had not moved, nor was any hardware loose, suggesting that there was at least a stable, growing union between the graft and the spine. The patient had an

excellent clinical outcome which we attributed to a functional fusion between C1-2, that was maintained throughout her early follow-up period. At 18-months postoperatively, dynamic cervical X-Rays demonstrated no hardware loosening, no subluxation of C1 on C2 and partial fusion of the bone graft to C1 and C2. Given these radiographic results, we are considering technical modifications including the use of mini-plates or cables to compress the O-VBG against the C1 posterior arch with placement of more allograft in between the two bony surfaces. 3. Discussion Fusion rates after C1-2 arthrodesis range from 75 to 97% for posterior wiring techniques [3] and 98% for screw-rod constructs [4]. Pseudoarthrosis, is rare after C1-2 fusion with screw-rod constructs, but presents a challenging problem for spine surgeons. The C1-2 complex is a highly mobile motion segment, particularly during axial rotation [5], and rigid surgical fixation is essential to promote bony fusion. Previous studies have explored the use of vascularized rib and scapular bone grafts to augment occipitocervical and cervicothoracic spinal fusion constructs [6,7]. Sagher et al. [8] described the use of a non-vascularized autologous occipital bone to supplement atlantoaxial fusion, however, this graft had the limitations of a non-vascularized bone graft, and was used in patients undergoing initial C1-2 fusion. Our study is the first surgical report to describe utilization of a vascularized occipital bone graft for failed C1-2 fusion. We believe this method for harvesting an occipital-based VBG could have applications at the craniovertebral junction as well as in the subaxial cervical spine. Since only a small piece of occiput needs to be harvested to cover the C1-2 arthrodesis site, the craniectomy does not need to extend more than 1 cm rostral to the foramen magnum. Therefore, should this fusion fail, a rescue

Please cite this article as: E. M. Reece, A. Vedantam, S. Lee et al., Pedicled, vascularized occipital bone graft to supplement atlantoaxial arthrodesis for the treatment of pseudoarthrosis, Journal of Clinical Neuroscience, https://doi.org/10.1016/j.jocn.2019.04.014

E.M. Reece et al. / Journal of Clinical Neuroscience xxx (xxxx) xxx

surgery with placement of an occipital plate should still be feasible with fixation between the top of the craniectomy site and the external occipital protuberance. While numerous small case series have been published on the use of VBGs in patients with complex spine pathologies [1,9–14], nearly all report on the use of free fibula or iliac crest grafts [1,9,10,12,14]. Although free-transfer VBGs can be successful, several challenges have precluded their wider acceptance among spine surgeons as a treatment option. These procedures add donor site morbidity, increased operative time, and increased blood loss; hence most surgeons have largely been limited to an anterior or combined anterior-posterior approach in order to accommodate the identification of adequate recipient vessels for the required microvascular anastomoses [15]. These challenges significantly increase the technical demand and length of the procedure [1,9,11,12,14]. In the present report, the site of harvest of the graft and its close proximity to the area of fusion overcome many of the limitations of free-tissue transfers, as well as the technical challenges of more complex pedicled VBGs. The postoperative imaging does not demonstrate as robust a fusion between the O-VBG and C1 as we anticipated, but the fusion to C2 does appear solid. This report serves as proof of concept that such a VBG can be safely harvested directly from the occiput in a manner similar to that performed by neurosurgeons in many other posterior fossa procedures. Furthermore, the vascularized bone graft may provide a constant flow of autologous osteogenic cells to the decorticated recipient bones through, thereby enhancing fusion. 4. Conclusions This is the first reported use of a semispinalis capitis muscle pedicled, vascularized occipital bone graft for use with revision atlantoaxial arthrodesis. This novel surgical technique leverages the advantages of vascularized structural autograft without adding extensive time or morbidity as commonly observed with freetissue transfers. We demonstrate the clinical feasibility of such a grafting procedure in this report and believe it may be a useful adjunct in revision surgery for atlantoaxial and perhaps occipitalcervical arthrodesis. Sources of support None.

5

IRB Approval for this study was obtained from the Baylor College of Medicine IRB. Declaration of interests None related to the submitted report. Acknowledgement The authors would like to thank Scott Holmes, CMI, a member of the Michael E. DeBakey Department of Surgery Research Core at Baylor College of Medicine, for his graphic assistance during the preparation of this manuscript. References [1] Wilden JA, Moran SL, Dekutoski MB, et al. Results of vascularized rib grafts in complex spinal reconstruction. J Bone Joint Surg Am 2006;88(4):832–9. [2] Lewis SJ, Kulkarni AG, Rampersaud YR, et al. Posterior column reconstruction with autologous rib graft after en bloc tumor excision. Spine (Phila Pa 1976) 2012;37(4):346–50. [3] Menendez JA, Wright NM. Techniques of posterior C1–C2 stabilization. Neurosurgery 2007;60(Supp 1):S103–11. [4] Elliott RE, Tanweer O, Boah A, et al. Atlantoaxial fusion with screw-rod constructs: meta-analysis and review of literature. World Neurosurg 2014;81 (2):411–21. [5] Lopez AJ, Scheer JK, Leibl KE, et al. Anatomy and biomechanics of the craniovertebral junction. Neurosurg Focus 2015;38(4):E2. [6] Menezes AH. Craniocervical fusions in children. J Neurosurg Pediatr 2012;9 (6):573–85. [7] Tubbs RS, Wartmann CT, Louis Jr RG, et al. Use of the scapular spine in lumbar fusion procedures: cadaveric feasibility study. Laboratory investigation. J Neurosurg Spine 2007;7(5):554–7. [8] Sagher O, Malik JM, Lee JH, et al. Fusion with occipital bone for atlantoaxial instability: technical note. Neurosurgery 1993;33(5):926–8. discussion 8-9. [9] Shin AY, Dekutoski MB. The role of vascularized bone grafts in spine surgery. Orthop Clin North Am 2007;38(1):61–72. [10] Shaffer JW, Davy DT, Field GA, et al. The superiority of vascularized compared to nonvascularized rib grafts in spine surgery shown by biological and physical methods. Spine (Phila Pa 1976) 1988;13(10):1150–4. [11] Shaffer JW, Davy DT, Field GA, et al. Temporal analysis of vascularized and nonvascularized rib grafts in canine spine surgery. Spine (Phila Pa 1976) 1989;14(7):727–32. [12] Ackerman DB, Rose PS, Moran SL, et al. The results of vascularized-free fibular grafts in complex spinal reconstruction. J Spinal Disord Tech 2011;24 (3):170–6. [13] Yelizarov VG, Minachenko VK, Gerasimov OR, et al. Vascularized bone flaps for thoracolumbar spinal fusion. Ann Plast Surg 1993;31(6):532–8. [14] Davis JB, Taylor AN. Muscle pedicle bone grafts; experimental study. AMA Arch Surg 1952;65(2):330–6. [15] Medgyesi S. Healing of muscle-pedicle bone grafts: an experimental study. Acta Orthop Scand 1965;35:294–9.

Permission Copyright permission from Baylor College of Medicine for the graphic illustrations appearing in the manuscript.

Please cite this article as: E. M. Reece, A. Vedantam, S. Lee et al., Pedicled, vascularized occipital bone graft to supplement atlantoaxial arthrodesis for the treatment of pseudoarthrosis, Journal of Clinical Neuroscience, https://doi.org/10.1016/j.jocn.2019.04.014