Laminoplasty : an evaluation of 24 cases.
Cervical expansive laminoplasty has been advocated as an alternative procedure to laminectomy for the decompression of the cervical spine. It provides favourable cord decompression and stabilisation of the cervical spine and is a simpler and safer alternative to anterior fusion and laminectomy for myelopathy and myeloradiculopathy, due to multisegmental cervical spondylosis and ossified posterior longitudinal ligament. We report our experience in 24 patients with this procedure, 12 of whom had myelopathy and another 12 had myeloradiculopathy. The earliest symptom to improve was radicular pain or paraesthesia (75%). A reduction in spasticity was seen in 21 of the 24 patients (87.5%). Eleven patients had improvement in their motor power during a follow up period ranging from 1 month to 14 months. One patient deteriorated following the procedure and developed Brown Sequard features due to under riding of the lamina on the hinged side, another had severe post operative paraesthesias, while one patient had a CT scan evidence of 'closing of the door', without being symptomatic for it. The technique of the procedure is discussed and the pertinent literature reviewed.
Expansive laminoplasty has been advocated as an alternative procedure to cervical laminectomy for patients with cervical myelopathy due to cervical canal stenosis, multisegmental cervical spondylotic myelopathy (CSM) and ossification of posterior longitudinal ligament (OPLL).,, It has the advantage over the latter procedure of avoiding instability and delayed deformity of the cervical spine, preventing the formation of laminectomy membrane and consequent cord recompression, and retarding the postoperative growth of OPLL which occurs due to instability of the cervical spine.,,,,, Several types of laminoplasties have been described.,,,,, We report a series of 26 patients treated by expansive open door laminoplasty. The technique of the procedure and a review of various types of procedures is discussed.
Over a period of 18 months, from August 1997 to January 1999, 26 laminoplasties were done at the National Institute of Mental Health and Neurosciences, Bangalore. In 24 cases it was done for cervical cord decompression and in the rest two for decompression of the thoracic cord. The former 24 cases form the basis of this study. The age of the patients ranged from 23 to 72 years with the majority being in the fifth decade [Table I]. The presenting signs and symptoms were as shown in [Table II]. 12 patients presented with myeloradiculopathy, while the other 12 patients had myelopathy alone. In the myeloradiculopathy group, 8 patients had radicular pain or paraesthesia and 9 patients had hypoaesthesias and sluggish reflexes in the radicular distribution. History of trauma as a precipitating symptom was forthcoming in 5 patients. The motor symptoms and signs predominated over the sensory, suggesting a primarily anterior compression. While virtually all patients had progressive weakness and stiffness of both lower limbs, only 15 cases had ascending hypoaesthesia and 12 had loss of joint and position sense. Eighteen patients were ambulant on presentation, with power of 4/5 in the lower limbs, while 6 had a power of 3/5 or less, and were not ambulant. Majority of the patients were in Nurick grade II or III (n = 18) [Table III].
All patients were investigated with X-ray cervical spine and plain CT scan (n=2), myelo CT (n=6), or with MRI (n=15), the latter being the preferred mode of investigation as it also showed the cord changes (hyperintense signals within the cord on T2W images). In one patient, myelogram was done because of logistic reasons. Sixteen patients had multisegmental cervical spondylotic myelopathy as seen with the above investigations, while 8 had OPLL. Four patients in the latter group had overt fluorosis with sclerosis of vertebrae, calcification of ligaments and of the interosseus membrane of the forearm, etc., as seen on plain X-rays.
Indications for laminoplasty included multisegmental cervical spondylotic myelopathy or myeloradiculopathy affecting three or more levels, with secondary canal stenosis, especially if high cervical region was involved, which is difficult to decompress and stabilise anteriorly, and OPLL of the continuous type. There was involvement of 3 levels in 8 patients, 4 levels in 9 patients, more than 5 level involvement in 2 patients.
Technique of laminoplasty : C3 to C7 spinous processes and laminae were exposed through a midline incision. The tips of the spinous processes were removed in chunks for use later as bone spacers. Drill holes were made in the remnant spinous processes for passing sutures. Using a drill, a gutter
was then created first on the side of the laminar elevation, at the lamina - facet junction. Both the outer and inner cortex were drilled on this side and using the 1 mm upcutting Kerrison's rongeur, the ligamentum flavum was removed on that side. Then lamina - facet junction of the other side was also drilled, sparing the inner cortex. Nonabsorbable sutures were passed through the base of the spinous processes and the laminae were elevated en bloc to create a gap of about 10 - 15 mm [Figure. 1]. The bone spacers were impacted in the gap. The side of laminar elevation was usually the left side which was closer to the surgeon, or on the side of the radicular symptoms so that a foraminotomy could be added to decompress the root canal. The sutures, through the spinous processes, were passed through the facet joint capsule and tightened so as to maintain the gap created [Figure - 2]. The wound was closed in layers.
Laminoplasty involving three laminae was done in 2 patients, four laminae in 6 patients and five or more laminae in 16 patients.
The earliest symptom to improve was radicular pain. Of the 8 patients with radicular pain or paraesthesia, 6 had improved in the immediate post operative period. In one patient, the paraesthesia worsened and was severe enough to warrant medication. It, however, improved on carbamazepine administration. In the postoperative period and follow up ranging from 1 month to 14 months, motor improvement was seen in 21 out of 24 patients, while 2 patients remained status quo. The earliest sign of motor improvement was a reduction in spasticity which was seen in all 21 patients in the postoperative period. Seven patients also had improvement in their motor power by the time of discharge, while another 4 patients showed improvement in motor power during the follow up period. While 17 of the 21 patients, who had improved, were ambulant, 4 remained nonambulant
and dependent for their activities of daily living, in spite of motor improvement. The improvement was by one grade (Nurick's grading) in 19 and by two grades in 2 patients. One patient deteriorated following surgery and developed postoperative Brown Sequard features with a right sided hemiplegia and left sided spinothalamic involvement. CT scan revealed under riding of the lamina on the hinged side at C3 level pressing over the cord. He was re-explored and laminectomy was done, following which the power improved to grade 4/5 proximally and 3/5 distally.
The complications encountered, apart from the neurological deterioration in one patient and severe post operative paraesthesia in another, included wound dehiscence in three patients because of gross spasticity and tension on the suture line. Two patients had wound infection. 'Closing of the door' was seen in one patient on follow up CT scan after one year, but he was not symptomatic.
Surgical decompression of the cervical spinal cord may be done through anterior approach or posterior approach. In the absence of prospective randomised trials comparing the two approaches, it is difficult to state a preferred choice of surgical treatment. Anterior decompression procedures include those that remove the involved vertebrae and discs as a block, and those that remove the disc at each level and replace it with a bone graft.,, These have the disadvantage of requiring prolonged operating time and retraction with consequent risk of oesophageal injury or carotid artery ischaemia for a three or four level decompression, especially in the elderly.,, The long segment strut grafts carry a potential risk of dislodgement and breakage due to osteopaenia in the elderly. Posterior approaches, in general, are preferred for patients with more than three levels of cord compression, elderly patients with poor bone quality and stenosis involving high cervical levels.,
Among the posterior approaches, laminectomy has remained the gold standard for comparison of the other procedures. However, the results of laminectomy have not been universally good. Post laminectomy spinal deformity has been reported in 10-15 % of patients undergoing laminectomy for cervical spondylosis,, and may occur due to several factors, the most important being lack of mechanical support, especially in the elderly patients with poor muscular tone, contributing to the weakening of the muscular support of the cervical spine. Yonenobu et al reported the development of instability and cervical kyphosis in 4 of their 22 patients after extensive decompression. Crandall and Gregorius noted late neurological deterioration after extensive laminectomy for cervical myelopathy in 60% of patients, as opposed to 20% undergoing anterior cervical fusion. However, this has not been such a common occurrence in older patients with cervical spondylotic disease, in other studies.
To offset the problem of post laminectomy deformity, cervical laminoplasty as a decompressive procedure of the cervical spine was developed by the Japanese. The procedure was originally described by Omaya and Hattori in 1973 as an expansive lamina Z plasty, and subsequently modified by Hirabayashi as an 'expansive open door laminoplasty'. Since then several modifications of laminoplasty have been described e.g. use of wire by Itoh, use of iliac bone graft, use of rib allografts in open door laminoplasty by Lee et al, use of midlongitudinal bone graft by Kurokawa et al, tension band laminoplasty, vascularised pedicled laminoplasty etc. We have followed Hirabayashi's technique in our cases, as other techniques take greater time and are technically difficult.
Several authors have documented the superiority of laminoplasty over laminectomy as a decompressive procedure. It has been claimed to reduce the postlaminectomy deformity. Laminoplasty done for myelopathy has been reported to demonstrate significant improvement in neurological functions,,,, and also significant relief from radicular pain, when done for multilevel spondylotic radiculopathy.
Various studies have correlated the clinical recovery with the amount of widening of the spinal canal. Hirabayashi demonstrated the best clinical results with a canal expansion of 4 to 5 mm sagittally. Kimura et al showed that a laminar elevation of 10 mm will expand the sagittal diameter of the spinal canal by 4 to 5 mm. Lee et al measured the canal body ratio and found an increase in the canal body ratio from 0.632 preoperatively to 0.871 postoperatively, by making an opening ranging from 12 mm to 18 mm. They found an improvement in gait disturbance in 84% (21 out of 25 patients), hand numbness and tingling in 87% and an improvement in bladder and bowel function in 77% of patients. Radiculopathy, when present, was alleviated in all when the procedure was combined with posterior foraminotomies. In our series, 6 out of 8 patients (75%) had an improvement in radicular symptoms, while 21 out of 24 patients (87.5%) had an improvement in spasticity and 11 out of 24 patients (46%) had improvement in motor power. Aita et al found that the mean sagittal diameter of the spinal cord at C5 increased by 0.8 mm, and the transverse diameter decreased by 0.9 mm after cervical laminoplasty. The mean cross sectional area of the cord increased by 7.4% and the centre of the spinal cord moved a mean 2.8 mm posteriorly at this level. It was postulated that the posterior movement may be the limiting factor in determining the decompressive effect of laminoplasty. Baba et al used 3D spiral computed tomography (CT) to assess canal widening and canal volume after laminoplasty and found that postoperative neurological improvement correlated significantly with increase in canal volume (of 45% in their series), and suggested that spiral 3D CT scan may be useful in evaluating the canal enlargement and prognosticating the outcome. Kohno et al, in a 5 year followup study of 22 patients, showed maximum good recovery to correlate with an optimal diameter increase of 5 mm. Even a small enlargement of the spinal canal results in improvement, which suggests that the improvement may not only be due to canal enlargement, but also because of improvement in the circulation of the spinal cord and nerve roots. It has been reported that the post operative growth of the continuous and mixed type of OPLL was 89% in the laminectomy group, 34% in the laminoplasty group and 20% in the anterior spinal fusion group, over a follow up period of over 5 years.
The commonest complication of this procedure is 'closing of the door' and restenosis. This can be obviated by suturing the lamina to the facet joint capsule, by impacting bone spacers (iliac crest or rib allografts, ceramic implants, etc.), or by using titanium miniplate stabilisation. Disjointing of the hinge resulting in floating lamina may also result in closing of the laminar door. Loss of cervical motion, especially rotation and lateral flexion can occur due to bony ankylosis along the hinge side of the laminoplasty. This, however, has the advantage of eliminating the postoperatve dynamic factor and helping stabilisation., Other complications are transient C5-6 root paresis and severe neck pain and paresthesias resulting from the tethering effect of the C5 and C6 nerve roots, as a result of the posterior movement of the dural tube and its contents. Under riding of the lamina at the hinged side occurred in one of our patients who developed a Brown Sequard syndrome and had to undergo a laminectomy.
Various predictors of outcome have been enumerated in several studies. Hirabayashi et al found the following causative factors to contribute to incomplete recovery : age over 65 years, severe disability in activity of daily living (less than 7 points on JOA score), myelopathy for more than two years, traumatic onset, severe stenosis of the spinal canal (more than 60%) and kyphotic curvature of the spine. Similar factors were proposed by other workers.,, Kohno et al found that cord degeneration and atrophy (depicted as a hyperintensity in T2WI on MRI) was associated with a less than optimal recovery, especially if multiple segments were involved. Lee et al, on the other hand, found no correlation between preoperative spinal cord signal change and the outcome. While identifying similar prognostic indicators as enumerated by Hirabayashi, they however found that even in patients having these negative predictors, more than 70% experienced post operative improvement in gait.
In summary, the expansive open door laminoplasty as described above, provides favourable cord decompression and stabilisation of the cervical spine and is a simpler and safer alternative to anterior fusion and laminectomy, for myelopathy or radiculopathy due to MCS, cervical canal stenosis, or OPLL. It requires no special instrumentation except for a drill, and no separate incision for harvesting a graft.