Cervical spondylotic myelopathy (CSM) refers to a clinical chronic disorder that is usually releated with degenerative disease of intervertebral disk (1). Cervical spondylotic myelopathy is the most common spinal cord degeneration in older sufferers, caused by the progressive spinal canal stenosis and subsequent nerve root compression (2). Surgical management is generally indicated for patients with CSM when conservative treatments are ineffective. Anterior cervical decompression and fusion for multilevel CSM is a complex procedure and may be associated with a long operative time, as well as complications, such as dysphagia, internal graft dislocation, and trigeminal nerve palsy (2). Laminoplasty (LP) and laminectomy (LC) with or without fusion are the primary posterior cervical surgical strategies for treating CSM to remove compressive elements, providing enough space for the cord, and decompressing the spinal cord (3).

Laminectomy, which is usually supplemented by additional fusion, was initially viewed as the gold standard practice for CSM (4). However, this technique is associated with many disadvantages, such as post-LC kyphosis, segmental instability, and subsequent neurological deterioration, which lead to a shorted indication. Laminoplasty was first reported by Tsuji et al. (5) in 1982, and is regarded as an effective way of maintaining anatomical cervical reduction. Laminoplasty retains a covering of the ligamentum flavum over the spinal cord and posterior laminar bone. Laminoplasty has the advantages of minimizing instability, limiting constriction of the dura from extradural scar formation, preserving motion, and avoiding complications related to fusion. However, LP is contraindicated in patients with CSM and >13° of kyphosis and severe neck pain (6). Although there are several disadvantages of LP, including vertebral canal reclosed problems, hinge fracture, higher technical requirements, and possible injuries to the cervical cord, LP has been gradually accepted by an increasing number of surgeons. At present, both surgical procedures decompress the spinal cord by enlarging the spinal canal and are regularly thought to be effective in treating CSM.

Although several meta-analyses or systematic reviews have been conducted to compare LC with LP in treating CSM, the results of these studies were not consistent. In a systematic review, Chen et al. (7) reported that LC with fusion had a higher rate of reoperation, non-union, and infection compared with LP. However, Fehlings et al. (8) suggested that LP and LC with fusion had similar effectiveness. Laminoplasty and LC with fusion may result in clinical recovery and a similar loss of lordosis. Similar to LC followed by fusion, expansive LP has a shorter operative time and less C5 palsy. In this study, we aimed to provide some references for clinical surgical treatment of CSM by systematically comparing the safety and efficacy of LP and LC regarding surgical outcomes, radiographic outcomes, clinical outcomes, and surgical complications.

Our study showed that patients receiving LP for CSM had less frequent occurrence of C5 radiculopathy and superficial infection than those who underwent LP. Length of operation, loss of blood, length of hospital stays, Cobb angle, SVA, VAS score, CCI score, Nurick score, NDI score, and other surgical complications were not different observed between these two groups. These results suggest that LP is a useful therapeutic procedure promoting management of CSM.

Several previous systematic reviews and meta-analyses that analyzed LP and LC with or without fusion for CSM have been published and showed different results compared with our study (26, 27). A meta-analysis by Lee et al. (28) compared LP with LC for treating CSM. In this article, a total of seven studies were included in the meta-analysis (six English papers and one Chinese paper). The authors focused on the clinical and radiological outcomes between these two different methods. The two groups did not show significant differences in JOA grade, VAS score, and CCI at the baseline state. The authors suggest that both methods may obtain clinical improvement and lead to a similar loss of lordosis, but definitive conclusion could not be reached regarding which surgical approach is more effective for the treatment of CSM. Liu et al. (29) presented a meta-analysis of 23 studies comparing LP with LC for treating CSM. They focused on clinical outcome (JOA, CCI, VAS, and cervical lordosis), complication (C5 palsy and axial pain), blood loss, and operation time. The LP group showed shorter operation time and fewer C5 palsy. Others had may achieve clinical improvement and a similar result. Phan and Scherman et al. (30) compared LP with LC for treating CSM in 10 studies when treating patients with CSM. They focused on the Postoperative JOA, postoperative VAS neck pain, postoperative CCI, postoperative Nurich grade, complication (reoperation rate and nerve palsy), operative time, and intraoperative blood loss. They found that there was no difference in terms of clinical improvement. However, a higher complication of nerve palsy was found in the LC than in the LC. Otherwise, results of others studies showed that LP and LC fusion methods were similarly useful (31).

Patients were matched and both groups had a similar length of operation, blood loss, and hospital stay. These characteristics were not examined in other relevant meta-analyses. Heterogeneity in our meta-analysis was high (I² > 75%), those personal difference and surgical processes had multivariate analyses and reports for them are speculative. Pooled results from three studies (21, 22) showed that the Cobb angle was more acceptable in the LP than in the LC. The reason why the Cobb angle was more acceptable in the LP group may be that the muscles and ligamentous structures of the cervical spine are dissected minimally and restored maximally in the LP group. In our study, clinical measurements, such as VAS, CCI, SF-36 MCS, SF-36 PCS, Nurick, and neck disability index scores, were not significantly different in two groups. However, the LP procedure was superior to the LC procedure in evaluation of the JOA score for patients with CSM. These findings indicated that the effect of LP was more favorable than that for LC, which suggested that LP could be considered as the method of surgery for patients with CSM. Most importantly, these results are supportive of a significant improvement in the management of CSM. Specifically, LP as a treatment strategy for CSM can obtain better results in the surgical process, radiographic outcomes, and clinical outcomes (Cobb angle, JOA score, and risk of C5 radiculopathy) than LC.

Several limitations must be considered when interpreting the results. First, the quality of the studies included 3 RCTs and 16 observational studies was low. Most of the RCTs did not provide any insufficient information on the exact methods of randomization to determine if credibility analysis was present. Allocation concealment was performed in three studied using sealed envelopes. Based on the selection criteria, we could not make further deletions or additions to the included papers. Second, the number of individuals in these studies were relatively small and therefore statistical power might be limited. Third, we guess that the research heterogeneity is mainly caused by the low quality of the included articles, which is mainly reflected in the high heterogeneity of the articles. The heterogeneity among the indicators included in our meta-analysis was mainly reflected in the time of operative, intraoperative blood loss, the time of hospital stay, ROM, SVA, Cobb angle, JOA, VAS, CCI, Nurick NDI, and Pseudarthrosis, and for the above heterogeneity, we used a random effects model for treatment statistics, while rate of C5 radiculopathy, Superficial infection, Hardware failure, Adjacent segment degeneration, Dural tea, Deep infection, Dysphagia, New radiculopathy (not C5), Postoperative kyphosis, and Neck/arm pain had lower heterogeneity and we used a fixed effects model for the treatment statistics. In addition, this general clinical heterogeneity may be caused by individual differences in patients, technical differences in the surgeon team, differences in medical equipment, and follow-up time. We considered whether our results might be affected by confounding factors. Therefore, high-quality RCTs are required to examine the long-term effects of these two surgical procedures on patients with CSM.

Our study shows that LP can achieve better results in C5 radiculopathy and superficial infection in surgical treatment of CSM compared with LC. These results suggest that LP is a therapeutic procedure for promoting management of CSM. Furthermore, high-quality research, adequately powered randomized studies are required to provide more evidence for the optimal surgical treatment of CSM for definite conclusions.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

ZZ and YZ: protocol and project development. LP: data collection or management. WM and SX: data analysis. HZ and RR: manuscript writing and editing. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.