Historical Development

In 1543, the anatomist and surgeon Andreas Vesalius (1514-1564) was the first to describe the intervertebral disc. However, its role in the development and as cause for various clinical signs and symptoms was not recognized until about 80 years ago. In 1928, Stookey³ reported on several clinical syndromes evoked by herniated cervical discs but interpreted the prolapse as “chondroma or neoplasm of the notochord.” Primarily, investigations by Schmorl⁴ in Europe and Keyes and Compere⁵ in the United States contributed to a detailed understanding of the pathophysiology of the intervertebral disc. Shortly after these investigations, Mixter and Barr⁶ clearly associated a lumbar disc prolapse with the signs and symptoms of nerve root compression; the alterations of the cervical intervertebral discs were identified as reason for neck pain, radiculopathy, and myelopathy.

Historically, the routine surgical approach for symptoms produced by cervical disc disease was posterior. During the last half of the last century, it became clear that the main problem of this approach was the technical challenge of displaying and removing compressive structures that lie anterior to the spinal cord and nerve roots. This need for easier access to reach anterior compressing structures led to the development of an anterior surgical approach to the cervical spine. Bailey and Badgley⁷ performed the first anterior stabilization of the cervical spine on a lytic process in 1952 and published their technique in 1960. In 1955, Robinson and Smith⁸ describe their method to stabilize a pathologically changed cervical spine segment using a horseshoe-shaped iliac bone graft. Without knowledge of the results of the other investigators, Cloward⁹ published his technique of anterior discectomy for removal of compressing structures in 1958.

Since that time, a multitude of variations and modifications have been done in the treatment of degenerative cervical disc disease. Different approaches are competing, as well as numerous implants, varying mainly in material and design. However, the crucial question of “to fuse or not to fuse” has not yet been answered. It is still a controversial issue if a favorable outcome can be achieved by decompression followed by solid fusion or if microsurgical decompression or decompression with arthroplasty can provide long-term preservation of motility.

When conservative therapy fails or progressive neurologic deficits are imminent, it comes to a decision for operative treatment. However, the decisions about what approach and what technique to use have to take a large number of variables into consideration. In the following section, the treatment evolution in operative management of cervical disc disease is illuminated and discussed, taking aspects such as approaches and techniques, indications, complications, advantages and disadvantages, and biomechanical consequences into account.

Laminectomy and Laminoplasty

Initially, the routine surgical approach for symptoms produced by cervical disc disease was posterior. Mixter and Barr⁶ published their report, which described the use of laminectomy to treat 19 cases of intervertebral disc herniation, including four ruptured cervical discs, in 1934. Since then, posterior approaches and techniques evolved from laminectomy into small keyhole foraminotomy and various laminoplasty methods.

The main indication for laminectomy in degenerative cervical disc disease is single- or multilevel stenosis of the spinal canal caused by various types of pathology, including spondylosis and ossification of the posterior longitudinal ligament, usually presenting with myelopathy or radiculomyelopathy. Laminectomy is preferably employed for patients over 65 years of age who demonstrate variable ventral pathology but with multilevel spondylosis and stenosis in the presence of a well-preserved cervical lordotic curvature. In addition, a medial facetectomy and foraminotomy can be performed at the necessary levels to address a specific disc herniation and spur.¹⁰

Open posterior approaches to the cervical spine with laminectomy avoid the approach-related complications associated with anterior approaches but require extensive subperiosteal stripping of the paraspinal musculature, which often results in significant postoperative pain. Moreover, accelerated degeneration of adjacent motion segments after anterior fusion, or so-called adjacent segment disease (ASD), is avoided.¹¹ Nevertheless, apart from anatomic limitations in reaching centrally or paracentrally located soft discs from this approach, long-segment decompression in patients who experience a preoperative loss of lordosis increases the risk for postoperative sagittal plane deformity.¹²

Complications after laminectomy include a 0% to 10% incidence of spinal cord injury and up to a 12.8% frequency of root injury.¹⁰ Nerve root deficits, typically involving C5, are variously attributed to direct mechanical manipulation or to “tethering” of the root after rapid cord decompression and dorsal migration after laminectomy.¹³

Laminectomy is successful in 77% to 85% of patients, while 10% to 15% may acutely deteriorate and another 23% exhibit delayed deterioration over 10 or more years.¹⁴ Rodrigues¹⁵ analyzed 51 patients who underwent laminectomy and disc removal for the treatment of paramedian and posterolateral soft cervical disc herniation and noticed 96% total relief of pain, 76% motor improvement, and 63% sensory improvement.

The clinical long-term outcome after cervical laminectomy,¹⁰ as well as the biomechanical alterations induced by multilevel cervical laminectomy,¹⁶ have been analyzed frequently. Performing extensive multilevel laminectomies typically does not immediately destabilize an otherwise intact spine¹⁷ but rather results in denervation and atrophy of the posterior cervical muscles. In addition, surgery denervates and disturbs the facet joints and the loss of the posterior tension band increases the force on the anterior vertebral body, which in turn worsens the sagittal deformity and results in a kyphotic angulation.

Several strategies have been used to prevent postoperative sagittal deformity after multilevel laminectomies. One option involves posterior segmental instrumentation and fusion performed at the time of the initial surgery. This adds considerable time and morbidity to the operation,¹⁸ results in a substantial decrease in mobility, and subjects adjacent levels to larger stress and accelerated degeneration. Another option includes laminoplasty or osteoplastic laminotomy. Because of concerns over complications of nonfusion-related laminectomy (spinal deformity, instability, compression by “laminectomy membrane,” and late neurologic deterioration), laminoplasty was developed in Japan in the 1970s as an alternative to laminectomy.¹⁹ A variety of laminoplasty techniques have been described; in all, the laminae are preserved, but the size of the spinal canal is expanded because the freed or partially freed laminae are positioned more posteriorly (Fig. 151-1).

FIGURE 151-1 Postoperative three-dimensional computed tomography (CT) scan after four-level cervical laminoplasty.

In their critical review on cervical laminoplasty, Ratliff and Cooper¹⁹ found no difference in neurologic outcome based on different laminoplasty techniques or when laminoplasty is compared to laminectomy. Furthermore, there was no benefit to laminoplasty over laminectomy in adult patients in terms of spinal alignment, incidence of kyphotic deformity, and the development of the postlaminectomy membrane. Cervical range of motion (ROM) decreased substantially after laminoplasty, with progressive loss in long-term follow-up studies and final ROM similar to that seen in patients who had undergone laminectomy and fusion. This gradual laminoplasty-induced decrease in cervical ROM may allow for gradual compensation by adjacent segments, thus limiting adjacent-segment disease.

Some studies compare the long-term outcome after laminoplasty and anterior spinal fusion for cervical radiculomyelopathy due to soft disc herniation²⁰ and for multilevel myelopathy.²¹ Because the two procedures provided the same neurologic improvement, the risks of bone graft complication with the anterior approach must be weighed against the risks of chronic neck pain associated with laminoplasty for determining the best technique. Furthermore, the posterior technique in this context may be reserved for multilevel soft disc herniation in older and multimorbid individuals.

The feasibility of minimal invasive access multilevel laminectomy and laminoplasty techniques were first shown in cadaver models.^22–24 Both techniques showed a 43% expansion of the cross-sectional areas of the spinal canal. In a clinical study on 10 patients with degenerative compressive myelopathy treated with endoscopic partial laminectomy, Yabuki and Kikuchi²⁵ showed that the technique was safely practicable and that all patients experienced symptomatic improvement. Furthermore, due to less muscle trauma and preservation of much of the normal osteoligamentous anatomy of the cervical spine, the postoperative wound pain and the risk for postlaminectomy kyphosis was reduced. Perez-Cruet²² reported on four patients with minimally invasive cervical laminoplasty and satisfactory postoperative neurologic improvement. Technically feasible, the author noted technical difficulties associated with elevation of the lamina and insertion of bone grafts due to the limited approach.

Posterior Foraminotomy

The posterior foraminotomy was pioneered by Frykholm²⁶ and Spurling and Scoville²⁷ and later modified by Scoville et al.²⁸ Accessible pathologies are laterally located, soft disc herniation and foraminal stenosis (hard disc), usually presenting with radicular symptoms (Fig. 151-2). Posterior foraminotomy may be performed unilaterally at one or more levels, bilaterally at one or more levels, or in combination with a laminectomy or laminoplasty.¹⁰ The detailed technique has been described elsewhere.^10,29 A major benefit of posterior foraminotomy versus posterior and anterior techniques with fusion is the preservation of stability and mobility. The avoidance of approach- and graft-related complications, impending in anterior techniques, is another important advantage. Limitations are the removal of compressive structures that lie anterior to the spinal cord and nerve roots.

FIGURE 151-2 (A) Axial magnetic resonance image showing a soft disc prolapse, and (B) CT scan showing a foraminal stenosis, both suitable for posterior foraminotomy.

Numerous publications focus on the indications, advantages, and long-term outcome of patients with lateral cervical soft disc herniation or osteophytes (foraminal stenosis), treated with posterior (keyhole) foraminotomy.^30–33 Successful relief of radiculopathy symptoms are seen in 90% to 98%, with complication rates between 0% and 4%, and recurrence rates of 0% to 6%.

Limiting the degree of medial facetectomy to less than 50% prevents postoperative instability³⁴ (Fig. 151-3). Because fusion is not applied, concerns about associated morbidity and the later development of adjacent-level disease theoretically are relatively minor. Nevertheless, Clarke et al. found in 303 patients with cervical radiculopathy, who were treated with posterior foraminotomy between 1972 and 1992, development of symptomatic adjacent-level disease in 4.9% after a mean follow-up of 7.1 years.³⁵

FIGURE 151-3 (A) Pre- and (B) postoperative CT scan after posterior foraminotomy, depicting the amount of bone removal.

Posterior cervical microendoscopic foraminotomy (MEF) and discectomy are the endoscopic alternative of microsurgical “open” posterior foraminotomy. The alternative was developed to address cervical nerve root compression by direct visualization of pathologic findings while minimizing tissue destruction and exposure. Muscle and ligamentous attachments to the spine are preserved, thus decreasing postoperative pain and spasm and helping maintain long-term stability.³⁶ Initial cadaveric studies using this technique demonstrated that the average vertical and transverse diameters of the laminotomy defect were identical for the MEF or open technique.³⁷ The average amount of facet removed and the length of neural decompression were greater with MEF than with the open approach. Initial clinical results have been quite favorable for the MEF procedure.³⁸ Operative times, blood loss, length of hospitalization, and need for postoperative pain medications have all generally been reduced when MEF is used, compared with the open procedure, yielding equivalent clinical results.^11,39,40 However, MEF is technically demanding, has a steep learning curve, and may require training within a cadaver laboratory.³⁶

Anterior Discectomy, Arthrodesis, and Arthroplasty

Anterior cervical discectomy (ACD) for spinal cord or nerve root compression was first described independently by Cloward,⁹ and by Smith and Robinson⁴¹ more than 50 years ago. Evolved from posterior techniques due to limited anatomic access, the common feature of this technique is the substantial or complete removal of the disc. Recently, anterior cervical foraminotomy was introduced,^42,43 which reduces the extent of disc removal but still has the anterior approach. The anterior approach to the cervical spine generally implies several approach-related complications (recurrent laryngeal nerve palsy, dysphagia, and esophagus and hypopharynx perforation)^44–46 that do not exist in posterior approaches. In his systematic review using evidence-based medicine to identify the best techniques for anterior cervical nerve root decompression, Matz⁴⁷ concluded that ACD, ACDF, and arthroplasty are effective techniques for addressing surgical cervical radiculopathy. However, the controversial question in recent decades has been, “Does anything have to be inserted in the empty disc space after anterior decompression,^48,49 and if so, what is the ideal implant/cage material and design for that?”^50,51 More questions have arisen, focusing on required additional anterior instrumentation and whether it might not be better to replace arthrodesis by arthroplasty.