Laminectomy

Laminectomy was the first procedure developed to treat CSM. Laminectomy is effective in the treatment of multilevel cervical spondylosis, OPLL, and other pathologic conditions causing cervical stenosis. A partial facetectomy is performed with this procedure and can contribute to postoperative cervical instability and deformity.

Posterior fixation is often combined with laminectomy because it has been shown to prevent or treat instability associated with multilevel laminectomy. In cadaver models laminectomy has been shown to significantly increase spinal cord flexibility and lead to instability. This instability can have various long-term consequences, including spinal cord microtrauma. This can result in neurologic sequelae and a tendency to develop kyphosis. To maintain alignment and restore the stability that is disrupted by decompression with laminectomy and partial facetectomy, posterior instrumented fusion is often added. Although fusion slightly increases the morbidity associated with the laminectomy procedure, stabilization of the spine can arrest the progression of spondylosis and potential deformity at the affected levels.

Perez-Lopez and associates compared results for 19 patients treated with laminectomy with results for 17 patients treated with laminectomy and fusion for cervical myelopathy. Similar clinical outcomes were noted in both groups as judged by Nurick scores. There was a greater increase in postoperative kyphosis in those treated with laminectomy (24%) than in those undergoing laminectomy with fusion (7%), which suggests that fusion prevents postoperative instability. The potential advantages of fusion are not certain. Hamanishi and Tanaka compared outcomes for 35 patients who were treated with laminectomy and 34 patients who underwent laminectomy and fusion and noted few differences between the two groups. Clinical outcomes as judged from improvement in Japanese Orthopaedic Association (JOA) scale scores were similar in both groups (scores improved by 51% in both groups) after a mean follow-up of over 3 years. Kyphotic malalignment occurred in 17% of patients in the nonfusion group and 12% of those in the fusion group. Radiographic instability occurred in two patients who did not undergo fusion and in two patients who did. However, any conclusions regarding the differences between fusion and nonfusion groups in this study are limited, because fusion was performed only in patients who were determined to have instability before surgery. Thus, the study had an inherent selection bias that steered patients whose condition was more stable toward nonfusion and those with greater instability toward fusion.

Various methods of posterior instrumented fusion have been used successfully to stabilize and fuse the spine. In the past when the posterior elements were absent, such as after a laminectomy, facet wiring with bone grafting was used as an effective method to stabilize the cervical spine. Facet wiring requires removing cartilage from the facet joint, drilling a hole through the inferior facet joint through which to pass a braided titanium cable, and adding an autologous bone graft. Currently lateral mass screws are routinely used and, compared with traditional wiring techniques, provide more immediate, rigid stability and thereby promote fusion. Cervical pedicle screws have been shown to provide better fixation and stability than lateral mass screws. Preoperative computed tomography (CT) is recommended to help view the three-dimensional anatomy necessary for accurate pedicle screw placement. However, the use of instrumentation to stabilize the cervical spine after laminectomy increases the frequency of neurologic and vascular injury. Thus, the benefits of improved stability with posterior instrumented fusion must be weighed against the higher risk of neurologic and/or vascular injury.

Laminoplasty

Because of the potential complications of laminectomy and posterior instrumented fusion, laminoplasty was developed in Japan in the 1970s. Various techniques for laminoplasty have been devised, but they all involve preservation of the lamina and decompression of the spinal cord by partially or completely freeing the lamina and positioning it more posteriorly. It was thought that laminoplasty would reduce the number of complications associated with postoperative spinal instability and deformity because it preserves the posterior elements of the cervical spine.

A Z-plasty laminoplasty technique was first described by Oyama. In this procedure the spinous process is removed and the laminae are thinned out to the facet joint. A “Z” shaped incision is made into the thin laminae, along with lateral troughs on both sides. Thinning the laminae allows them to be manipulated and separated to expand the spinal canal. The manipulated segments of lamina are held in place with sutures or wire material ( Figure 6-4 ).

Original Z-plasty laminoplasty technique. The spinous processes are removed ( A ) and the laminae are then thinned ( B ). Troughs are drilled laterally ( C ). Finally, a Z-shaped cut is made between the adjacent thinned laminae ( D ). Here, the laminae are secured in an open position using sutures.

The expansive open-door laminoplasty was described by Hirayabashi and colleagues. A high-speed drill is used to drill down to the ligamentum flavum on one side of the lamina that has been exposed. The facet capsules should be left intact. A trough is created on the contralateral side at the same location. Unlike on the open side, the ligamentum flavum is not reached and a thin section of the lamina anterior cortex is left intact on this side. The lamina on the open side is carefully pulled away from the spinal cord. Distraction of the lamina posteriorly expands the spinal canal. The lamina can be maintained in a more posterior position with the assistance of sutures, titanium miniplates, or small bone graft struts ( Figure 6-5 ). Herkowitz combined this procedure with a unilateral foraminotomy on the open side of the laminoplasty and noted improvements in 90% of patients with cervical radiculopathy. Several variations of the open-door laminoplasty rely on instrumentation for internal fixation of the mobilized laminae ( Figure 6-6 ). Shaffrey and associates described the use of bone graft with titanium miniplates to hold the lamina in place. Gillett and colleagues described a device called a CG clip that serves as a spacer and an attachment unit to provide stabilization of the open laminae without the need for bone grafting.

Open-door laminoplasty. The spinous processes are removed ( A ) and bilateral troughs created at the facet-lamina junctions ( B ). A thin rim of bone is left on one side ( C ). The laminae are elevated and, in this modification, secured to the facet using sutures ( D ).

Open-door laminoplasty with hardware-augmented internal fixation. The techniques are similar to those in the Hirabayashi-type laminoplasty. The spinous processes are removed ( A ) and bilateral troughs created at the facet-lamina junctions ( B ). A thin rim of bone is left on one side ( C ). In the final step, as described by Shaffrey and colleagues, ²³ titanium miniplates are placed to secure the elevated laminae and hold the bone graft in place ( D ).

The French door laminoplasty or double-door laminoplasty was described by Kurokawa and co-workers. Troughs are created bilaterally just medial to the facet joints, and the spinous process is split at the midline using a thread wire saw (T-saw) or high-speed drill. The spinous process is thus split in half and the laminae are spread apart through the midline. The laminae are held in a more posterior position with sutures attaching them to the facet capsules and wiring or ceramic spacers placed between the split spinous process ( Figure 6-7 ). In a modification of this procedure, bone graft can be harvested from the spinous process and placed as a spacer. No clear advantage over the open-door technique has been shown with this procedure, but it has been proposed that splitting at the midline allows for a more symmetric decompression of the spine.

French door (double-door) laminoplasty. A, The spinous processes are split using a high-speed drill or T-saw. B, Troughs are drilled into the lateral aspects of the laminae and medial facets. C, The spinous processes are split in the midline. D, Wiring, bone graft, or ceramic material is used as a spacer.

Laminoplasty has become an increasingly popular technique for treating cervical spondylosis. As with laminectomy, several concerns and complications have been noted with this technique. Persistent axial neck pain, C5 paresis, and reduced range of motion have been reported. Hosono and colleagues observed a 60% prevalence of postoperative axial symptoms in a group of 72 patients treated for CSM. Takemitsu and associates reported a 14% rate of C5 palsy in 73 patients treated with laminoplasty for CSM. The rate was much higher in the 10 patients in whom laminoplasty was combined with posterior instrumentation to treat instability and kyphosis (50% rate of C5 palsy). Tanaka and colleagues noted transient C5 nerve palsy in 3 of 62 patients treated with laminoplasty despite intraoperative spinal cord monitoring with transcranial motor evoked potentials. Several studies have confirmed that laminoplasty is associated with reduced postoperative cervical range of motion. Chiba and co-workers followed 80 patients who underwent open-door laminoplasty for CSM or OPLL for an average of 14 years and noted a 36% reduction in cervical range of motion. Kawaguchi and colleagues followed 126 patients for longer than 10 years after laminoplasty for CSM. Range of motion decreased to 25% of preoperative range of motion. Although dural lacerations are more commonly associated with laminectomy, they have also been documented with double-door laminoplasty. Dickerman and associates noted delayed dural lacerations due to dislodgment of hydroxyapatite spacers in 4 of 130 patients treated with this procedure for CSM.

Skip Laminectomy

Axial neck and shoulder pain, restriction of cervical range of motion, and the reduction of cervical lordosis have been the significant postoperative problems noted with laminoplasties of the cervical spine. To minimize the damage to the posterior musculature, skip laminectomy was developed. In this procedure, a standard laminectomy at the affected levels is combined with a partial laminectomy of the cephalad half of the laminae at other levels, at which the muscular attachments to the spinous processes are left undisturbed. This technique was first described by Shiraishi in 2002.

The initial results of skip laminectomy have been very encouraging. In 2003, Shiraishi and associates published a retrospective review comparing skip laminectomy with open-door laminoplasty. The authors noted increased axial neck pain in 66% of the patients undergoing laminoplasty, whereas only one patient (2%) in the skip laminectomy group developed axial neck pain. The patients in the skip laminectomy group maintained 98% of their range of motion, whereas the patients in the laminoplasty group maintained only 61% of their range of motion.

In another study by Yukawa and associates , the authors conducted a prospective, randomized trial comparing laminoplasty and skip laminectomy. Forty-one patients with CSM were randomly assigned to undergo either a laminoplasty (n = 21) or skip laminectomy (n = 20) and were followed for longer than 1 year (average, 28.1 months). No significant differences were seen between groups in operative invasiveness, axial neck pain, cervical alignment, or range of motion.

Hence, skip laminectomy has shown some promise, but more studies with longer follow-up are necessary before its efficacy relative to conventional laminectomy or laminoplasty can be determined.