452

Novel Atlantoaxial Bone Graft Fixation Technique Koffie RM, et al. https://doi.org/10.14245/ns.1938344.172 660 www.e-neurospine.org a C1–2 bone graft. 7,8 With the evolution of modern instrumentation systems, sur- gical fixation using various screws and rods has demonstrated excellent clinical and radiographic fusion rates. Surgeons now have more tools to obtain arthrodesis at the craniovertebral junction with less need for noninstrumented wiring bone graft techniques with external orthoses. Magerl and Seemann 9 pro- posed using C1–2 transarticular screws for fixation in this re- gion. Goel and Harms then reported the use of C1 lateral mass screw placement in combination with C2 pars or pedicle screws with high fusion rates. 10,11 A variant of the Goel technique, in which an allograft spacer is placed within the C1–2 joint in ad- dition to bilateral C1 lateral mass and C2 pedicle or pars inter- articularis rods and screws fixation has also been described. 12 This approach results in higher fusion rates, but was associated with increased rates of postoperative occipital headaches and may increase the risk for vertebral artery injury. Ghostine et al. 13 also reported a similar approach in which arthrodesis along the entire C1–2 joint is used to complement instrumented fixation, but without sacrificing the C2 nerve roots, resulting in lower rates of occipital headaches in patients postoperatively. Another method of atlantoaxial fixation that obviates the need for C2 nerve root sacrifice or compression by using C2 translaminar screws in combination with C1 sublaminar cable suspension for C1–2 fusion in the setting of atlantoaxial instability has also been described. 14 Here we report an operative nuance, which combines the Goel-Harms fixation technique with a safe and facile, modified Sonntag arthrodesis technique for securing interlaminar bone graft between C1 and C2. Our variation entails wedging an iliac crest graft, either harvested autograft or structural allograft, be- tween C1 and C2 for interbody arthrodesis and anchoring it with a 0-Prolene suture. The curvature and low-profile nature of the needle as well as the easy handling of the 0-Prolene su- ture makes sublaminar passage under C1 relatively straightfor- ward. We demonstrate the effectiveness of this technique in a series of 32 patients with atlantoaxial stability for a variety of pathologies commonly encountered in practice and confirm excellent radiographic evidence of fusion at minimum 6 months and low complication rates. Illustrations and operative video highlight the technique. MATERIALS ANDMETHODS Institutional Review Board (IRB) approval at the Massachu- setts General Hospital (2015P001837) was obtained to perform a retrospective database query of a prospectively maintained spine surgery database developed by the senior author at our tertiary care academic institution. Individual patient consent was not collected given the retrospective nature of this case se- ries and technical report. Standing postoperative plain lateral and anteroposterior radiographs were obtained prior to dis- charge and at 6 weeks, 3 months, 6 months, 1 year, and 2 years postoperatively. Cervical spine computed tomography (CT) was obtained at 6 months after surgery to assess the graft posi- tion and the extent of arthrodesis. Fusion was radiographically confirmed by the following criteria: (1) absence of radiographic lucency between iliac crest bone graft and C1 inferior laminar and C2 superior laminar on thin cut CT; (2) absence of dynam- ic motion on plain lateral and anteroposterior x-rays; (3) pres- ence of new bone formation and remodeling across C1–2. Pa- tients were managed in a soft collar for comfort after surgery. No halo immobilization or rigid cervical orthoses were used in the postoperative setting. Thirty-two consecutive adult patients were identified who met the inclusion criteria for the study. Primary and metastatic tumor cases were excluded. Cases including occipital plate fixa- tion were excluded. Patients with less than 6 months of follow- up were excluded. Data were collected on hospital stay, read- mission rates, perioperative complications, patient-reported out- comes with Neck Disability Index (NDI) questionnaires, and fusion rates based on cervical CT scans. 1. Description of Technique The patient is positioned prone on the Jackson table and a standard approach to the C1–2 region is performed. C1 lateral mass and C2 pars interarticularis screws are placed using the Goel-Harms technique. After the instrumentation is placed in the C1 lateral masses and the C2 pars interarticularis, an angled curette is used to dissect soft tissue away from superior and in- ferior ledges of the C1 lamina. A plane is created safely under- neath the lamina so that the epidural space is free from attach- ment to the posterior C1 ring. Meticulous dissection is required to avoid neurologic injury and dural disruption. Dissection is continued until an adequate sublaminar corridor is obtained. For purposes of arthrodesis, either a structural iliac crest auto- graft or allograft can be used. In the accompanying video, iliac crest autograft is used. In many cases and in the illustrations provided, structural allograft is used. The bone graft is sized to fit within the C1–2 interlaminar space and decortication is performed on the inferior aspect the C1 lamina and the superior aspect of the C2 spinous process | The Surgical Technologist | AUGUST 2021 352