A US National Library of Medicine PubMed search was conducted for articles pertaining to cervical laminoplasty and fusion for the management of CSM. A PICOS—P (problem/patient/population), I (intervention/indicator), C (comparison), O (outcome), and S (setting)—worksheet and search strategy was created and followed (data not shown). Institutional review board approval was not required for a literature review.

The initial search items used included “cervical,” “laminoplasty,” “fusion,” and “myelopathy.” Article reference lists and suggested similar studies through PubMed were checked to identify additional pertinent studies. The search was limited to human studies, English-language literature, and reports on more than one patient. A total of 263 articles were initially identified (Fig. 2). A PRISMA (preferred reporting items for systematic reviews and meta-analyses), flow chart (an evidence-based minimum set of items for reporting in systematic reviews and metaanalyses) was then utilized (Fig. 2).

The criteria for final selection led by two of the authors (KJB and LNS) included patients with myelopathy, in which both cervical laminoplasty and fusion were performed together. Based on these criteria, six articles were included in the quantitative analysis, all of which were retrospective case series reviewing patient data between 1995 and 2009 (Fig. 2).

The principle summary measures included the Frankel grade (FG) of spinal cord injury; the Barthel Index, which measures functional status by assessing activities of daily living; the Japanese Orthopedic Association (JOA) Score to assess the clinical symptoms in a setting of CSM; the Visual Analog Scale for rating pain; and the C2–C7 angle to assess kyphosis. Data from each study were reviewed and placed in a table format to allow for comparative assessment of all records (Table 1) [7-12].

In 2003, Matsumura et al. [10] described a case series of two adult patients with multilevel CSM and spinal instability, who were treated with posterior stabilization using lateral mass plates and expanding laminoplasty by using interspinous iliac bone graft. One of these patients was a 76-year-old woman who presented with progressive motor weakness and sensory disturbances in both upper and lower extremities [10]. Her multilevel CSM was due to a kyphotic deformity with instability from C3 to C5, as demonstrated by significant motion on dynamic cervical radiographs. Magnetic resonance imaging (MRI) revealed degenerative spinal canal stenosis and spinal cord compression. The patient underwent laminoplasty with interspinous iliac bone graft and posterolateral mass plating from C3 to C5. The surgeons bent the lateral mass plates in the procedure to accommodate her cervical lordosis. Her deficits had improved by the time of discharge and she could perform her daily activities without any support by 3-week follow-up visit (Fig. 3). Complete bony fusion was radiographically evident at 15 months postoperatively. Three and a half years after her operation, she remained independent with no radiographic evidence of instability or re-stenosis.

In 2012, Uehara et al. [7] retrospectively reviewed 13 patients who had CSM and instability from 1998 to 2010, specifically excluding patients with trauma, rheumatoid arthritis, destructive spondyloarthropathy, spinal tumors, or athetoid cerebral palsy. Instability was defined by the White and Panjabi’s guidelines for radiographic instability, including vertebral body slip length of >3.5 mm and a vertebral body angle change of >11° between the adjacent vertebral bodies. However, if the patients’ symptoms were severe and spinal cord compression was mild, they underwent posterior instrumented fusion regardless of whether the radiographic measurements met the above criteria. Cervical pedicle screws were placed using a computed tomography-based navigation system, after which expansive open-door laminoplasty was performed. The JOA scoring system, which measures the motor and sensory functions, was used to evaluate CSM before surgery, immediately after surgery, and at the final follow-up, ranging from 12 to 83 months. The cervical sagittal alignment (C2–C7 lordotic angle) was compared at similar time periods. The results revealed that the JOA scores had significantly improved from 9.1±3.4 before surgery to 12.6±3.3 at the final follow-up (p=0.009) (Fig. 4). The average C2–C7 lordotic angle was 8.8°±11.0° preoperatively and 5.3°±10.9° (p=0.021) at the final follow-up (Fig. 5).

Miyamoto et al. [9] retrospectively reviewed 60 patients with CSM and kyphotic deformity who had undergone either laminoplasty alone or laminoplasty with posterior fusion at three local centers from 1999 to 2009. Thirty additional patents with CSM, but no kyphotic deformity, were used as controls and they underwent laminoplasty alone. As reported by Matsumura et al. [10], Kurokawa’s lamina splitting method was used for these patients, but the hydroxyapatite spacers were placed between the lamina, as opposed to that in the bone graft. Of the 30 patients who underwent combined laminoplasty and fusion, 18 received lateral mass screws and 12 received pedicle screws. The primary outcomes were the recovery rate of the JOA score, local kyphosis, and C2–C7 angle. The authors found that the recovery rate of the JOA score was significantly worse in the laminoplasty-only patients than in the laminoplasty and fusion patients (32.6% versus 44.5%, p<0.05). They a significant difference in the correction of local kyphotic angle with a mean correction angle of 11.1° postoperatively in the laminoplasty and fusion group, whereas the laminoplasty-only group maintained a local kyphotic angle (p<0.05). The C2–C7 angle was improved at the follow-up for the laminoplasty and fusion group (−0.4° to −11.1°), whereas it was maintained in the control group (−12.0° to −11.6°) and had worsened in the laminoplasty-only group (−1.7° to 0.5°, p<0.05) (Fig. 5). The authors concluded that improvement in patient outcomes after fusion was attributable to the improvement in cervical lordosis, improvement in posture, decompression of thickened posterior ligamentum flavum, and decompression of the area around the pre-surgical kyphotic deformity.

Segmental instability is an important pathological factor in CSM due to ossification of the posterior longitudinal ligament (OPLL). While laminoplasty has been widely used in CSM due to OPLL, progressive kyphosis and progressive ossified lesions on the follow-up remain as concerns, potentially leading to late neurological complications [15,16]. Chen et al. [12] reviewed a cohort of 15 patients with CSM due to OPLL who had dynamic instability and who were treated with unilateral open-door laminoplasty with titanium arch miniplate (no bone graft) and lateral mass screw fixation at unstable levels to provide immediate stabilization from 2005 to 2007. Using this surgical technique, none of the 15 patients developed progressive kyphosis or progression of the ossified lesions in over 4 years of the follow-up. Furthermore, the JOA scores significantly improved at 6 months postoperatively and well-maintained at the follow-up (mean, 7.5±1.6 points, preoperatively; 13.0±2.0 points, 4-years postoperatively) (Fig. 4). These patients experienced improvement in their neurological functions at a rate of 57.9% in addition to the improvement of neck pain at 4-years postoperatively. These authors compared their results with another series of 83 patients with CSM due to OPLL treated with laminectomy and instrumented fusion [14]. They found nearly identical neurological recovery; however, laminectomy and instrumented fusion was associated with a higher risk of complications such as nerve root palsy and hematoma. Thus, the authors concluded that laminoplasty with fusion is safer than laminectomy with fusion, although they have similar clinical effectiveness [12].

In acute subaxial traumatic central cord syndrome without fracture or dislocation, Song et al. [8] retrospectively reviewed the surgical management to evaluate the role of surgery in this patient population. Twenty-two patients were identified at a single institution from 1995 to 2002. Of the 22 patients presented, four received laminoplasty and fusion with lateral mass screws. This approach was used for patients with cervical spinal instability on dynamic radiographs and multilevel CSM or OPLL. The remainder of the patients received anterior decompression and fusion or laminoplasty alone. All of the patients who received laminoplasty and fusion presented after a fall (as opposed to a motor vehicle collision) and had pre-existing developmental canal stenosis. The interval between injury and surgery was 3 days for two patients, 6 days for one patient, and 37 days for another patient due to the compromise of pulmonary function associated with the injury. All of these patients improved neurologically by at least one FG at discharge (Fig. 3). Laminoplasty with lateral mass screw fusion provided rapid decompression and fixation with good results in patients, with unstable multilevel CSM or OPLL.

In the case series of Matsumura et al. [10] a 57-year-old construction worker with CSM, who fell from a 5-foot scaffolding, is described. Imaging demonstrated cervical canal stenosis from C3 to C7 with cord T2 hyperintense signal at both C3–C4 and C6–C7. In addition, a dynamic instability was noted between C6 and C7 in association with a fracture of the C6 facet joint extending into the lamina. The patient underwent posterolateral fixation from C5 to C7 with lateral mass plates and screws in addition to C3–C7 laminoplasty by the Kurokawa’s method, involving insertion of hydroxyapatite spacers between the split lamina of C3 and C4 and the insertion of iliac bone graft secured with titanium wire between the lamina from C5 to C7. This patient, who presented with moderate weakness, could ambulate independently at 1 month and achieved radiographic bony fusion at 15 months postoperatively.

Demura et al. [11] presented a retrospective case series at a single center on the management of CSM due to athetoid cerebral palsy using laminoplasty and pedicle screw fixation. Athetoid cerebral palsy is caused by perinatal cerebellar and basal ganglia damage and is characterized by involuntary movements of the neck, trunk, and extremities [16]. CSM occurs early in this patient population due to excessive mechanical stress caused by involuntary movements, which lead to cervical spine degeneration and instability. Patients with CSM, due to athetoid cerebral palsy, provide a unique treatment challenge because the involuntary neck movements increase the postoperative risk of kyphosis, pseudoarthrosis, and adjacent segment disease. A consecutive series of 15 patients underwent midline T-saw laminoplasty with spinous process splitting, followed by pedicle screw fixation and iliac bone grafts between the opened lamina and pedicle screws. Postoperatively, patients showed improvement of their sagittal alignment with C2–C7 angle improved from 11.0°±14.5° of kyphosis to 1.5°±12.7° (Fig. 5). The postoperative JOA scores had a mean improvement rate of 32% at the final follow-up (mean follow-up time, 71 months). There were two cases of adjacent segmental disease necessitating additional surgery; however, all patients achieved solid posterior bony fusion without needing rigid orthosis, and there were no neurovascular complications. This study concluded that pedicle screw fixation allowed for better biomechanical strength, while the laminoplasty opened a space for iliac bone grafting to facilitate fusion, which when combined, provided for strong stabilization. The patient demographic, associated pathology, procedure data, measure of neurological improvement, C2–C7 angle pre- and postoperatively, and follow-up from some previous studies are summarized in Table 1 [8-12].

This literature review aimed to explore the utility of laminoplasty in combination with instrumented fusion, and six articles, all of which were retrospective case series, were quantitatively analyzed. The present review was restricted to articles in the English literature; as such, some relevant articles may have been excluded. Retrospective reviews come with inherent bias and sources of error. Data were recruited by convenience sampling, such as performing chart reviews at our institution; thus, the results may not be representative of the general population and may be prone to selection bias. Meanwhile, confounding was also a possibility, as postoperative variables may have been distorted by the presence of other factors not considered during data collection. Moreover, as this was a retrospective study, causation was not determined, rather only associations were made. Rare scenarios such as CSM treated with both laminoplasty and posterior fusion require large sample sizes, whereas this retrospective institutional review was unlikely to yield this information. Lastly, wherever electronic records were utilized, potential bias was introduced from items copied and pasted, default tick boxes, and the use of boilerplate notes.