Cervical Laminoplasty

Published on 02/04/2015 by admin

Filed under Neurosurgery

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 2363 times

Chapter 46 Cervical Laminoplasty

General Principles and History

Laminectomy was first introduced to release the spinal cord compressed at multiple levels, although it fell into relative disfavor due to complications such as laminectomy membrane, segmental instability, kyphosis, and late neurologic deterioration. Ventral decompression and fusion or posterior fusion in addition to laminectomy was a solution in the United States and European countries, whereas laminoplasty was created in Japan, especially for treating ossification of the posterior longitudinal ligament (OPLL). Such ossification is difficult to remove directly via a ventral approach because the extremely hard ossification often tightly adheres to the dura mater. Direct resection of the ossification, therefore, was strongly associated with the potential risk of disastrous cord damage, and postoperative displacement of the grafted bone or pseudarthrosis was not rare, because a long bone graft was needed to span the trough after resection of the long OPLL. All of these complications kept most surgeons away from employing ventral surgery for cervical OPLL. Laminoplasty was developed as a safer and more reliable procedure to treat OPLL in 1971 by Hattori et al.,1 who expected to enlarge the spinal canal and to relieve neural compression while maintaining a skeletal and ligamentous dorsal arch to prevent epidural scarring and malalignment of the cervical spine. Although this procedure, the so-called Z-shaped laminoplasty, was rather complicated, simpler and more feasible laminoplasty procedures were devised and are now divided into two categories: unilateral (hinge) laminoplasty and bilateral (hinge) laminoplasty. Given that the patients with compressive myelopathy generally have a developmentally narrow spinal canal, decompression over the entire cervical spine with laminoplasty seems more reasonable than ventral decompression surgery, in which operated levels are restricted and adjacent segment disease can take place several years later. Thus, the number of patients with compressive myelopathy who undergo laminoplasty is increasing each year. Several trials to eliminate the disadvantages of laminoplasty are discussed herein.

Indications

The surgical indication for compressive cervical neuropathy is a myelopathy that progresses despite treatment. It is, however, difficult to determine when to apply surgery to cervical myelopathy, because its natural history remains obscure. Apparent ambulatory disturbance is a definite indication, but symptoms no more profound than finger numbness are debatable indicators. Some surgeons prefer the less popular prophylactic laminoplasty to prevent accidental spinal cord injury for patients with a narrow spinal canal even if they have only slight neurologic symptoms. It is, however, difficult to eradicate the risk of spinal cord injury by doing laminoplasty; some patients with OPLL who have residual cord compression after laminoplasty can sustain cord injury due to minor trauma.

Indications for laminoplasty should be discussed in contrast with those for ventral and other dorsal techniques. Generally speaking, a patient with spinal cord compression at one or two levels is a good candidate for anterior decompression and fusion unless the anteroposterior canal diameter is equal to or less than 13 mm. Because most patients with myelopathy who require surgical decompression have a developmentally narrow canal, they are candidates for laminoplasty. Although the spinal cord is assumed to migrate dorsally and escape from anterior lesions by laminoplasty, such a mechanism may not work in two special conditions: kyphosis and the presence of a large anterior lesion. In a kyphotic cervical spine, dorsal cord migration may not be expected after lamina opening, yet some surgeons argue that kyphosis of less than 5 or 10 degrees can benefit from laminoplasty. The extent of kyphosis for which laminoplasty can effectively release the spinal cord remains unknown. The spinal cord does not seem to escape from large and/or steep ventral lesions, even after sufficient dorsal space is provided by laminoplasty. Herniated nucleus pulposus, however, is successfully treated by laminoplasty. Neurologic improvement is excellent after laminoplasty for disc herniation, regardless of whether the herniated nucleus is absorbed after surgery. Beak-type OPLL, in contrast, does not seem to be successfully treated by laminoplasty. Resection or floating of the ossification via the anterior approach should be considered for these patients, although these methods are technically demanding and associated with a high rate of surgical morbidities.

Contraindications

A cervical kyphosis of greater than 5 to 10 degrees is considered a contraindication for laminoplasty, because the spinal cord cannot be released from the anterior lesion if the dorsal space is made by laminoplasty.

Elderly patients who tolerate general anesthesia may be candidates for laminoplasty because the operative impact of laminoplasty is acceptable. There exist, however, arguments regarding the operative outcome for elderly patients. Potential risks for postoperative delirium and cardiovascular accidents should be taken into account.

Subaxial lesions in rheumatoid arthritis (RA) have been treated with arthrodesis, although reduction of neck motion, swallowing disturbance, and adjacent segment disease are not rare after spinal fusion. Laminoplasty is an alternative to diminish the drawbacks associated with arthrodesis. Retrospective investigation in our series revealed that patients with nonmutilating-type RA can benefit from laminoplasty if subaxial subluxation is mild.2 In contrast, mutilating-type RA and/or RA with vertebral slippage more than 5 mm is a contraindication for laminoplasty. Cervical myelopathy associated with athetoid cerebral palsy may be best treated with laminoplasty combined with fusion. A screw-rod system or a long bone graft spanning all fused levels with a postoperative halo vest is a common technique to attain spinal fusion. Laminoplasty alone has little effect on the myelopathy of athetoid cerebral palsy. Patients undergoing hemodialysis may be candidates for laminoplasty, unless they have destructive spondyloarthropathy, in which spinal instability should be managed by spinal fusion. Pyoderma on the nape skin is a contraindication for laminoplasty, because of the high risk for surgical site infection. Pyoderma is an infectious dermal disease well observed on buttock skin, although head and neck regions may also be affected.

Although laminoplasty was originally developed to treat OPLL, occasional neurologic deterioration is reported immediately after laminoplasty for massive OPLL. The reason for this complication is unclear, but surgeons with expertise have a good reason for choosing ventral surgery for OPLL that has a thickness greater than 50% of the spinal canal.

Techniques

Various types of laminoplasty are in clinical use. They are divided into two major categories: unilateral (hinge) laminoplasty and bilateral (hinge) laminoplasty. In unilateral laminoplasty, or open-door procedure, two bony gutters are drilled on either side of the lamina-facet junction. The gutter on one side is cut out and the lamina is opened by elevating this edge, while the gutter on the other side functions as a hinge by following gentle fracture. The side to be opened does not depend on the laterality of compression. A left-side opening is generally convenient for right-handed surgeons. The opened lamina is kept in situ by sutures placed between holes drilled in the lamina and the facet joint capsule. Postoperative reclosure of the lamina, however, can take place, and the opening space may be spanned by a spacer to maintain the enlarged spinal canal. Resected spinous processes or ceramic spacers are often inserted at every two laminae and fixed by sutures between the lamina edge and the lateral mass. The nonfixed laminae are also kept open by a yellow ligament attached to the adjacent fixed laminae (Fig. 46-1). Small metal plates are alternative implants to maintain the opened lamina, although they are not as popular in Japan as in Western countries. Metal plating adds to the complexity of the operation, is time-consuming, and adds to the expense.

With bilateral laminoplasty, or the double-door (French door) procedure, three bony gutters are drilled not only on either side of the lamina-facet junction but also in the midst of the spinous process. After the midline cut is made, each half of the lamina is opened laterally, similar to opening French doors. The lamina was originally kept in situ by inserting a bone graft between each half of the lamina; at present the most popular insertion materials are ceramic spacers. Although ceramic spacers are usually fixed by sutures, they often become displaced in the early postoperative period. The more dorsally the spacers are inserted, the more often the spacers become displaced.

Although the superiority of unilateral or bilateral laminoplasty has been discussed, significant differences between them have not been found so far. Intraoperative blood loss, operating time, outcome, and morbidities are all supposed to be similar between the two kinds of laminoplasties. One more bone gutter to be made in the midst in bilateral laminoplasty seems to be time-consuming for surgeons who prefer unilateral laminoplasty, whereas occasional epidural bleeding from the open side gutter in unilateral laminoplasty seems troublesome for surgeons who advocate bilateral laminoplasty.

When radiculopathy accompanies myelopathy, nerve roots can be released by foraminotomy in addition to laminoplasty. In unilateral laminoplasty, foraminotomy facilitates nerve root exposure on the open side. In bilateral laminoplasty, aggressive foraminotomy might destroy the bony gutter and result in lamina separation. Much care should be taken not to violate the bony gutters. Microsurgical foraminotomy is an alternative method of releasing the nerve roots.

Electrophysiologic monitoring with somatosensory-evoked potentials, motor-evoked potentials, and electromyography is not mandatory for laminoplasty. Inadvertent neural injury cannot be avoided by intraoperative monitoring, and laminoplasty is a relatively safe procedure. Arguments exist over whether electrophysiologic monitoring can detect complications such as C5 palsy. This is doubtful.

During the introduction period of laminoplasty in Japan, a cervical collar was generally applied for a few months after surgery. Surgeons thought that external support was a prerequisite to facilitate bony union of the hinged gutters or grafted bones. However, as unfavorable spine fusion and aggravation of axial neck pain were recognized as the adverse effects of collar application, many surgeons discontinued this practice. In contrast, patients are encouraged to perform isotonic muscle exercises in the early postoperative period to prevent muscle weakness.

Outcomes

The operating time for laminoplasty is 1 to 3 hours and intraoperative blood loss is 100 to 500 mL. Allogenic transfusion is not usually required. Neurologic gain after laminoplasty is generally excellent. The recovery rate averages from 55% to 60% when the patients are evaluated by the JOA (Japan Orthopedic Association) score, which has subsets for motor and sensory functions of the extremities and bladder function. The operative outcomes, however, vary depending on the outcome measures, including the JOA score, Nurick’s scale, and various kinds of performance tests. The grip-and-release test that counts finger motion cycles in maximum effort in 10 or 15 seconds reveals 100% recovery after laminoplasty. It is unclear that laminoplasty produces superior neurologic outcome over laminectomy or anterior surgery, because few studies have conducted a randomized controlled comparison of these surgical procedures. Factors that can predict the outcome of laminoplasty include the patient’s age, period of disease, preoperative neurologic status, transverse area of the spinal cord, and preoperative signal changes on MRI. Low-intensity changes on T1-weighted images and high-intensity changes spanning multiple levels on T2-weighted images are indicators of poor functional recovery. The outcome of laminoplasty for large OPLL is considered to be especially poor when the anteroposterior occupying ratio to the spinal canal is more than 60%.

Long-term outcomes of laminoplasty are also excellent and maintained for 5 years from surgery, after which neurologic gain is gradually lost in some patients. Approximately 30% of the patients who underwent laminoplasty were reported to encounter neurologic deterioration in the 10-year follow-up.4 Late deterioration is more frequent in cases of OPLL (27–30%) than in cases of spondylosis (16–30%).4 Neurologic deterioration after laminoplasty can be attributed to osteoarthritis of the hip or knee joints, degenerative lumbar diseases, cardiovascular and cerebrovascular diseases, minor trauma, age-related dysfunction, progression of cervical OPLL, and thoracic spine ossification, although no causative factors can be identified in some patients. Increase in OPLL thickness is as frequent as 70% of cases in the 10-year follow-up after laminoplasty, although neurologic deterioration results in only 3% to 7%.5 Thus, the long-term outcome of laminoplasty can be concluded to surpass that of anterior surgery, which definitely has adjacent segment diseases.

Complications

Perioperative Complications

One of the most disastrous complications of laminoplasty is spinal cord injury. In drilling bony gutters with a high-speed drill, it is possible for the bur on the drill to injure the dura mater or even the spinal cord. If the bleeding is such that a clear field cannot be maintained in the drilling area, the bur can go dangerously deep. Because the loss of resistance method is not reliable for preventing inadvertent perforation of the lamina, every effort should be made to minimize bleeding so the drilling area is easy to visualize. Bleeding from the bone marrow of the pedicle, which often makes the irrigation water opaque, may be addressed by applying bone wax. A surgical microscope is not required for the common type of laminoplasty, as long as a clear visual field is maintained by the measures described here. Spinal cord injury with laminoplasty is rare compared with that in the anterior surgery, in which the visual field and working space are smaller than those in laminoplasty.

Another cause of spinal cord injury is the stepwise decompression nature of the procedure. If the laminae are drilled and opened one by one, the decompressed spinal cord migrates dorsally with each opening. The spinal cord can be kinked at the edge of the residual lamina, causing cord injury. Laminae should therefore be opened en bloc over the total decompression area.

Epidural bleeding is a common complication of spinal surgery. The typical blood loss for laminoplasty is about 100 to 500 mL. The epidural venous plexus is rich in the lateral part of the spinal canal, but it is sparse in the midline. In unilateral laminoplasty, bony gutters are made just on the rich epidural vein and bleeding can be massive. Surgeons should take care not to perforate the inner cortex of the lamina, especially when using a steel bur. The best way to cut off the lamina on the opened side is to crack the evenly thinned lamina by rotating the elevator inserted in the gutter (Fig. 46-2). Even after every lamina is cut off in this way, massive bleeding can occur during lamina opening. If a suction tube is effective to allow visualization of the dural surface, lamina opening should be continued to the last lamina prepared because dural expansion by lamina opening often squeezes dilated veins. If bleeding is overwhelming and far beyond suction capacity, the opening process should be transiently interrupted and surgeons may have to wait for a few minutes after placing hemostatic collagen onto the vein until the blood flow decreases. Huge vein networks are difficult to cauterize using a bipolar coagulator. Bleeding, however, is minimal in most cases of unilateral laminoplasty.

Epidural hematoma formation is another common complication of spinal surgery. A subfascial closed wound drain tube should be placed to prevent hematoma formation, although the tube placed on laminae often has no effect in evacuating blood collection under laminae (Fig. 46-3). Laminoplasty has an advantage in that an opened lamina functions as a protector against posterior muscles that may compress the spinal cord, whereas with laminectomy, the exposed spinal cord is susceptible to compression by hematoma or muscles.

Wound dehiscence can occur, especially in patients with athetoid cerebral palsy, in whom dense sutures with nonabsorbable material are recommended to approximate the fascia. Surgical site infection is not frequent after laminoplasty, except in patients with infectious skin diseases such as pyoderma. When the dura mater is injured, most often by a high-speed drill, cerebrospinal fluid leaks postoperatively. The fascia should be sutured densely in a water-tight manner if dural laceration is recognized during surgery. Other treatments for cerebrospinal fluid leak are discussed in another chapter.

Postoperative displacement of an implanted ceramic spacer is more often observed after bilateral laminoplasty than after unilateral laminoplasty (Fig. 46-4). Although ceramic spacers usually migrate dorsally, they can cause not only dural laceration but also cord injury if they displace ventrally.6 Lamina dropping, or falling forward, on the hinged side is one of the most common complications of laminoplasty. When surgeons realize the complete separation of the hinged side cortex, they should remove the floating lamina to avoid neural injury. Making the bony hinge appropriately flexible is a critical point of laminoplasty, because too loose a hinge makes the lamina drop and too rigid a hinge results in reclosure of the opened lamina (Fig. 46-5).

Palsy of C5 has been the biggest topic of debate in cervical spine surgery. Scoville7 and Stoops8 described this complication after laminectomy in 1961, but it was first reported after laminoplasty in 1986. Postoperative C5 palsy is defined as paresis of the deltoid muscle and/or the biceps brachii muscle after cervical decompression surgery without any deterioration of myelopathy symptoms. The vast majority of C5 palsies occur within a week following surgery, and recent studies reveal a shorter latency between surgery and the onset of the C5 palsy than had been previously considered. Some patients present with palsy on the day of surgery. Although the reason why the palsy occurs exclusively in the unilateral C5 nerve root has been extensively discussed, recent papers reveal that the palsy occurs in every root, including C5, C6, C7, and C8, individually or in combination.9,10 The incidence of palsy is 5% for only the C5 root but 10% for all roots. The incidence of palsy is similar between laminoplasty and anterior cervical surgery. C5 palsy generally recovers spontaneously as long as the palsy is mild. Most mild palsy with a manual muscle testing (MMT) grade of 3 or 4 fully recovers within 6 months, whereas severe palsy with an MMT grade 2 or less recovers only up to a useful level, often taking more than 6 months.

The cause for C5 palsy or upper limb palsy in a broader sense still remains unknown. Inadvertent injury to the nerve root during surgery, nerve root traction caused by consecutive dorsal shifting of the cord following decompression surgery (“tethering phenomenon”), spinal cord ischemia due to decreased blood supply from radicular arteries, segmental spinal cord disorder, and reperfusion injury of the spinal cord have been proposed so far, although none of these alone can effectively account for all of the clinical characteristics of C5 palsy. Tethering phenomenon has been considered the most likely pathogenesis of C5 palsy for a long time. Some authors, however, report that C5 palsy does not necessarily emerge in patients whose dorsal migration of the spinal cord is excessive after laminoplasty.11,12 Spontaneous recovery of C5 palsy or the palsy after anterior surgery cannot be accounted for by this tethering theory. Reperfusion injury has been advocated recently as a possible cause of C5 palsy. A chronically compressed spinal cord may be injured by free radicals after being exposed to rapid reperfusion of the blood flow. However, the distribution of palsy restricted to a single segment is difficult to explain by reperfusion of the spinal cord. The most recent hypothesis for C5 palsy is thermal damage to the nerve roots. The experimental data suggest that tissues adjacent to drilled bone, especially nerve roots, can be damaged by friction heat from a high-speed drill, which is often beyond 100°C without water irrigation.13 In experiments simulating hyperthermia therapy, not only the latent period between thermal damage and palsy but also motor recovery after several weeks is indicated. This characteristic coincides with the clinical course of C5 palsy.

Axial neck pain is the most frequent complication of laminoplasty, with an incidence of 10% to 20%. It can be defined as the appearance of neck and shoulder pain after cervical spine surgery. Since its first report in 1992, many attempts were made to reduce this notorious complication. The most effective preventive measure is to discard the cervical collar after surgery. Long-term collar application definitely aggravates axial pain, and no collar application decreases the incidence and intensity of axial pain. A trend of not using a collar after laminoplasty has emerged. The pathogenesis of axial pain, however, is still obscure. Because axial pain is much more frequently observed after laminoplasty than after anterior surgery, the disruption of posterior neck tissues is suspected to be the origin of pain. Intermittent decompression by skip laminectomy14 or minimally invasive laminoplasty using a tubular retractor seems promising to decrease axial neck pain by preserving posterior neck tissues. Such procedures, however, appear complicated and time-consuming. We realized that postoperative axial pain significantly decreased by limiting the range of laminoplasty from C3-7 to C3-6.15 The C7 spinous process with various tissue attachments has a critical biomechanical importance and should be spared from the range of laminoplasty. Given the rarity of cord compression at C6-7 in cervical spondylosis, C3-6 laminoplasty seems to be a necessary and sufficient procedure, although another strategy may be needed in treating OPLL.

References

1. Kawai S., Sunago K., Doi K., et al. Cervical laminoplasty (Hattori’s method). Procedure and follow-up results. Spine (Phila Pa 1976). 1988;13:1245-1250.

2. Mukai Y., Hosono N., Sakaura H., et al. Laminoplasty for cervical myelopathy caused by subaxial lesions in rheumatoid arthritis. J Neurosurg. 2004;100(Suppl 1):7-12.

3. Yonenobu K., Wada E., Ono K. Laminoplasty for myelopathy. Indications, results, outcome and complications. In: Clark C.R., editor. The cervical spine. ed 4. Philadelphia: Lippincott Williams & Wilkins; 2004:1057-1071.

4. Chiba K., Ogawa Y., Ishii K., et al. Long-term results of expansive open-door laminoplasty for cervical myelopathy: average 14-year follow-up study. Spine (Phila Pa 1976). 2006;31:2998-3005.

5. Iwasaki M., Kawaguchi Y., Kimura T., Yonenobu K. Long-term results of expansive laminoplasty for ossification of the posterior longitudinal ligament of the cervical spine: more than 10 years follow up. J Neurosurg. 2002;96(Suppl 2):180-189.

6. Kaito T., Hosono N., Makino T., et al. Postoperative displacement of hydroxyapatite spacers implanted during double-door laminoplasty. J Neurosurg Spine. 2009;10:551-556.

7. Scoville W.B. Cervical spondylosis treated by bilateral facetectomy and laminectomy. J Neurosurg. 1961;18:423-428.

8. Stoops W.L. Neural complication of cervical spondylosis; their response to laminectomy and foraminotomy. J Neurosurg. 1961;19:986-999.

9. Sakaura H., Hosono N., Mukai Y., et al. Segmental motor paralysis after cervical laminoplasty. A prospective study. Spine (Phila Pa 1976). 2006;31:2684-2688.

10. Hasegawa K., Homma T., Chiba Y., et al. Upper extremity palsy following cervical decompression surgery results from a transient spinal cord lesion. Spine (Phila Pa 1976). 2007;32:E197-E202.

11. Kurosa Y., Yamaura I., Nakai O. Pathophysiology of postoperative C5 nerve root palsy [in Japanese]. Spine and Spinal Cord. 1993;6:107-114.

12. Sodeyama T., Goto S., Mochizuki M., et al. Effect of decompression enlargement laminoplasty for posterior shifting of the spinal cord. Spine (Phila Pa 1976). 1999;24:1527-1531.

13. Hosono N., Miwa T., Mukai Y., et al. Potential risk of thermal damage to nerve roots by a high-speed drill—a possible cause of C5 palsy after cervical spine surgery. J Bone Joint Surg [Br]. 2009;91:1541-1544.

14. Shiraishi T. A new technique for exposure of the cervical spine laminae. Technical note. J Neurosurg. 2002;96(Suppl 1):122-126.

15. Hosono N., Sakaura H., Mukai Y., et al. C3-6 laminoplasty takes over C3-7 laminoplasty with significantly lower incidence of axial neck pain. Eur Spine J. 2006;15:1375-1379.