The optimal surgical treatment option for multilevel cervical OPLL remains controversial. Surgical options include anterior corpectomy and fusion, laminoplasty, and laminectomy and fusion.
Anterior corpectomy and fusion allows for direct removal of the OPLL mass and is more effective at restoring cervical lordosis than posterior surgery. However, disadvantages of anterior approach are technical difficulty and high complication rates which include pseudarthrosis, dysphagia, and dural tears. So the posterior approaches are preferable with more than three levels involved.
The common posterior approaches, laminoplasty and laminectomy and fusion, use an indirect decompression with less technically demanding and lower rate of complications.
Laminoplasty accomplishes decompression by hinging open the laminae and results in a 30% to 40% increase in the size of canal volume. Compared with laminectomy and fusion, Laminoplasty is advocated because of its preservation of neck range of motion (ROM). 
However, neck ROM may incite further progression of OPLL. Progression of the ossification was found in 66% of the patients underwent laminoplasty. According to Yoshida, limiting cervical ROM may prevent late deterioration due to progression of OPLL. Morio reported that restriction of segmental motion was associated with clinical improvement in myelopathy.
In addition, complications associated with laminoplasty include closure of the opened lamina, hinge fracture, development of postoperative malalignment. Significant kyphosis and instability are contraindications for laminoplasty.
So laminectomy and fusion may be preferred for the treatment of multilevel cervical OPLL. Laminectomy and fusion removes the laminae followed by instrumented fusion and results in a 70% to 80% increase in spinal canal. According to Houten, laminectomy and posterior lateral mass fusion can lead to high rates of fusion, preserved lordosis, and clinical results comparable or superior to those seen with anterior surgery.
A significant portion of laminectomy requires posterior fusions caudally to C6 or C7. However, cervicothoracic junction is a transition point between the lordosis of the cervical spine and the kyphosis of the thoracic spine. Furthermore, the subaxial cervical spine provides up to 20° of combined flexion/extension, 10° of lateral bending, and 5° to 7° of rotation per level. This mobility is in stark contrast to the structurally rigid thoracic spine, which permits <5° of flexion/extension and lateral bending per level. The substantial difference between mobility in cervical and thoracic spine may amplify rates of adjacent segment disease at the cervicothoracic junction when multilevel cervical fusions are terminated in the lower cervical spine.[16, 17]
So, routine extension of posterior cervical fusions into the thoracic spine has been suggested. The benefits of extension into the thoracic spine include greater surface area for the fusion mass and the larger screws typically employed in the thoracic spine which may offer greater construct rigidity and a more stable mechanical environment.
According to Osterhoff et al, secondary interventions due to adjacent segmental pathology or implant failure were necessary in 18/58 (31.8 %) of the C7-cases and in 1/16 (6.3 %) of the T1/2-cases (p=0.038). So, they suggested that patients with multi-segmental posterior cervical fusions ending at C7 showed a higher rate of clinically symptomatic pathologies at the adjacent level below the instrumentation. One may consider to bridge the cervico-thoracic junction and to end the instrumentation at T1 or T2 in those cases.
As Schroeder et al presented, a significant difference in the revision rates was identified between fusions terminating at C7, T1, and T2-T4 (35.3%, 18.3%, and 40.0%, P=0.008). Patients whose construct terminated at C7 were 2.29 (1.16–4.61) times more likely to require a revision than patients whose construct terminated at T1 (P=0.02), but no difference between stopping at T1 and T2-T4 was identified. So, they recommended that multilevel posterior cervical fusions should be extended to T1, as stopping a long construct at C7 increases the rate of revision.
In our study, grafted bone was completely fused in all patients. There was no adjacent segment disease and instrument failure occurred during the follow-up.
Progression of OPLL could be observed both during the natural course and after surgery. The incidence of postoperative progression reported in the literature varied from 3.3% to 74.5%. As Sakai K et al presented, postoperative progression of OPLL was observed in 5% of the anterior decompression and fusion with floating method group and 50% of the laminoplasty group. Lee et al reported that the incidence progression of OPLL was 45.5%, 62.5%, and 30% for laminoplasty, laminectomy, and laminectomy with fusion, respectively. Lee et al performed a meta-analysis of 11 studies and reported a 62.5% incidence of OPLL progression after laminoplasty and 7.6% after anterior or posterior fusion surgery. So, posterior decompression with instrumented fusion surgery could suppress the progression of OPLL. In our study, progression of OPLL was observed in two patients(4.3%) at final follow-up.
C5 palsy is a serious complication after cervical decompression surgery in which the patient shows deterioration in power of the deltoid or biceps brachii. According to Pan FM et al, the average incidence rate of C5 palsy after posterior cervical spine surgery was 7.8% (range, 1.4–23.0%). Risk factors for C5 palsy included age, male gender, OPLL, and stenosis of the C4–C5 intervertebral foramen. Foraminotomy and intraoperative neuromonitoring were the two main methods used to prevent C5 palsy. In our study, both foraminotomy and intraoperative neuromonitoring was used to prevent C5 palsy.
Axial symptoms are defined as pain from the nuchal to the periscapular or shoulder region. According to Wang M et al, The pooled axial symptoms prevalence was 28 % (95 % CI 24–32). The prevalence of axial symptoms was higher after expansive open-door laminoplasty (39 %) than after modified open-door laminoplasty (23 %) and laminectomy instrumented fusion (29 %). They suggested that postoperative axial symptoms may be reduced through preserving posterior muscles and structures, stabilizing cervical vertebrae, and reducing external cervical immobilization time. The semispinalis cervicis, most of which inserts on C2, acts as a dynamic stabilizer and extensor of the cervical spine. For complete preservation of the semispinalis cervicis inserted in C2, Takeuchi et al changed the laminoplastic procedure from C3–C7 laminoplasty to C4–C7 laminoplasty with C3 laminectomy. Their report demonstrated that modified laminoplasty with preservation of the semispinalis cervicis significantly reduced postoperative axial symptoms. In our study, muscle insertion of C2 spinous process was retained and cervical collar was worn for one month, which were performed to prevent axial symptoms.
C2-C7 lordosis angle and C2-C7 SVA were used to measure the translation of cervical spine in the sagittal plane. Lee et al reported expansive laminoplasty vs laminectomy alone vs laminectomy and fusion for cervical OPLL. Cervical lordosis, C2-C7 Cobb angle and CCI, decreased gradually in all patients. SVA was maintained in laminectomy and fusion group only and increased in the others. According to Liu X et al, the SVA significantly increased in expansive open-door laminoplasty and was maintained in laminectomy and instrumented fusion for cases with cervical OPLL. C2-C7 Cobb angle increased in laminectomy and instrumented fusion and decreased in expansive open-door laminoplasty. Our results showed that C2-C7 Cobb angle increased and C2-C7 SVA decreased, which was similar with the result of Liu X et al.
JOA was used to evaluate neurological recovery. Excellent neurological recovery is associated with the extent of decompression. Extensive laminectomy, as an alternative surgical option, allowed adequate decompression of the spinal cord and nerve roots by removing spinous processes, lamina, ligamentum flavum, and especially the inner less than 1/4 of facet joints on each side. According to Du W et al, enlarged laminectomy with fixation for the management of multilevel cervical degenerative myelopathy was demonstrated to be an effective strategy for improving neurological function. We performed enlarged laminectomy in patients with multilevel cervical OPLL and excellent neurological recovery was obtained.
There are several limitations in this study. Firstly, it is a retrospective, single-institution study. Then, there is no control group and sample size is small. Finally, follow-up time is insufficient. So, randomized controlled studies with long-term follow-up are needed to confirm these results.