Guest Discussants: Hugh Bassewitz, MD, Jon Kimball, MD, Langston T. Holly, MD
SpineLine Section Editor: Jeffrey C. Wang, MD

Case Presentation
A 68-year-old Japanese woman visiting from Japan, who had cervical surgery 20 years ago, is now complaining of decreased fine motor control and dexterity. Other significant problems include balance and bilateral numbness. She is hyperreflexic and has a positive Hoffmann’s sign. She has no neck pain.

If this patient wanted surgical treatment, would you recommend an anterior or posterior approach?

Hugh Bassewitz, MD, Responds
The first issue for me is to ensure that this woman needs surgery. A CT-myelogram would be a must for me before planning the definitive operation, as T2-weighted MRIs can exaggerate stenosis.

Her myelopathic symptoms could arise from what appears to be long-standing cord deformity and atrophy including the myelomalacia behind C5-6. She has significant adjacent level degeneration with stenosis and cord deformity at C4-5 and less at C6-7 (although C6-7 looks fairly preserved on the axial MRI).1 Reasonable amounts of cerebrospinal fluid (CSF) surround the cord both anteriorly and posteriorly on all three axial cuts. I would counsel that surgery could be undertaken to prevent further progression of her myelopathy. She should understand that, even with surgery, her symptoms could progress. Decompression is not likely to eliminate her symptoms entirely.

The principles of surgery in this patient would focus on eliminating any residual stenosis and restoring cervical lordosis.2 The anterior approach achieves both of these goals. The C4-5 level would be best treated with anterior discectomy, as there is obvious stenosis and cord deformity. I would probably also include the C6-7 level. Two separate transverse incisions afford excellent exposure and cosmetically give optimal results. Because she has had prior anterior cervical surgery, I would need to examine the incision and perhaps use it for the current approach.

The Smith-Robinson anterior discectomy and fusion is one of the most successful spinal operations. There are many variations on the surgery, but a direct anterior decompression of C4-5 with an anterior reconstruction with either iliac crest autograft or allograft and plating has an excellent track record. A high fusion rate and restoration of cervical lordosis can be achieved through positioning, distraction and a lordotic graft. The same procedure could be undertaken at C6-7.

To avoid dissection through scar, an approach from the contralateral side could be undertaken. If a contralateral approach is selected, an indirect laryngoscopy is useful to assess function of both recurrent laryngeal nerves and that their respective vocal cords have normal function. If the previously operated side demonstrates poor vocal cord function, an ipsilateral approach through the scar prevents a potential devastating injury to the other recurrent laryngeal nerve. An oropharyngeal gastric tube can be used to help identify and protect the esophagus during the approach.

The C5-6 level presents the toughest challenge in this case. If surgical goals include eliminating all lesions, a C5-6 decompression could be considered due to the presence of stenosis, cord deformity and the clinical symptoms of myelopathy. Three reasons to ignore C5-6 include the previous procedure at that level, its myelomalacia, and the presence of a ring of CSF around the cord. If C5-6 decompression is required, C5 and C6 corpectomies could be performed for maximal decompression. This approach has disadvantages including the difficulty dissecting through scar and the increased blood loss noted with corpectomies. On the other hand, the advantages include: total elimination of any stenosis and complete removal of disc and bone causing her stenosis. Structural iliac crest autograft, fibular strut allograft, titanium or PEEK cages can be used. When cages are employed, the patient’s own harvested bone can be packed inside. I prefer a static plate that allows some settling through screw angulation rather than a true translational plate.

In this case, a posterior-only approach would require a multilevel posterior cervical laminectomy and fusion. This operation would effectively decompress the cord, but would not restore the patient’s lordosis. Additionally, if the patient remains kyphotic, fixation at C7 would be subjected to significant pullout forces. Therefore, C7 and probably T1 pedicle screws would be indicated, which can be challenging. Also, infection rates are higher in posterior cervical spine surgery.

For all of these above reasons, I would choose either anterior cervical discectomy and fusion at C4-5 and C6-7 or a  wo-level corpectomy and fusion at C5 and C6. For the corpectomy option, I would choose a stacked PEEK cage construct, filled with autogenous corpectomy bone, using caspar posts and head halter traction to restore lordosis before cage placement. I would place a cervical plate from C4 to C7, and use a rigid collar postoperatively for three months. With this operation, one can effectively treat the anterior column stenosis and restore the cervical lordosis.3

References
1. Kienapfel H, Koller M, Hinder D, et al. Integrated outcome assessment after anterior cervical diskectomy and fusion. Spine. 2004;29:2501-2509.

2. Faciszewski T, Jensen R, Hand C. Recurrent central cord syndrome at the level of a solid cervical vertebral fusion. Spine. 2003;28:E1-79-E182.

3. Sasso R, Ruggiero R, Reilly T, et al. Early reconstruction failures after multilevel cervical corpectomy. Spine. 2003;28:140-142.

Jon Kimball, MD, and Langston T. Holly, MD, Respond
The success of surgical intervention in cervical spondylotic myelopathy (CSM) depends on a comprehensive evaluation of the patient’s individual clinical and radiographic findings. This is especially true in revision scenarios in which the surgeon must consider the initial treatments and their sequelae. Disagreement exists in the literature regarding the management of CSM and timing of surgery. Most studies looking at operative treatment are retrospective in nature and nonoperative treatment has been shown to have similar outcomes in elderly patients with  mild myelopathy. However, the results of operative treatment and opportunity for neurologic recovery are generally better in patients who undergo decompression earlier.1,2

The primary goal of surgical intervention of patients with CSM is expansion of the spinal canal and decompression of the spinal cord. This can be accomplished by anterior, posterior or combined approaches. Each allows varying degrees of deformity correction and stabilization depending upon the patient’s alignment, morphology, disease severity and previous operations. When working up a revision cervical patient, we find that static MRI and standard AP and lateral plain films are often insufficient. Whenever possible, we prefer to evaluate the dynamic alignment of the considered levels with flexion extension radiographs. A CT scan is also helpful in decision-making whenever there is question about the presence of ossification of the posterior longitudinal ligament or disc calcification. The patient’s history and symptoms are equally important. Determine how quickly her symptoms have progressed. Does she have Lhermitte’s sign? Does she have radicular symptoms from foraminal impingement? Swallowing difficulty or voice changes?

In this case, a 68-year-old Japanese woman with a remote history of noninstrumented anterior cervical C5-6 fusion presents with myelopathy without radicular symptoms from adjacent segment degeneration both above and below the fused C5-6 levels. There is a prominent disc protrusion at C4-5 with effacement of the spinal cord. There is also residual stenosis and flattening of the cord at the fused C5-6 level. The sagittal MRI demonstrates only mild to moderate C6-7 stenosis, but this level is adjacent to a previous fusion and there may be a component of dynamic stenosis. Additionally, there appears to be T2 weighted spinal cord signal change at each of the aforementioned levels. Thus, this patient’s case raises four primary questions: (1) is it necessary to operate? (2) If so, which spinal levels require decompression? (3) What is the optimal approach to treat her degenerative and congenital stenosis? (4) Given the revision nature of the surgery, what risk factors and comorbidities specific to this patient will most likely affect her outcome?

The answer to the first question is clearly yes. Although some of her long tract signs may be related to her initial presentation and management 20 years ago, the recent onset hand and gait problems are classic for myelopathy. Based on the fact that there is significant canal narrowing, spinal cord flattening, likely dynamic stenosis and T2 weighted signal change between the C4-5 to C6-7 levels, each of these levels require decompression. Overall, the patient has a mildly kyphotic spinal alignment, which causes some surgeons to reflexively rule out a posterior approach and opt for an anterior procedure. In general, anterior decompression and fusion procedures are preferable in myelopathic patients with significant kyphosis or large ventral compressive lesions causing spinal cord deformation. In those situations, posterior decompression will not adequately disengage the cord anteriorly. In that setting, posterior approaches could potentially yield a worsened neurological condition and poor outcome. On the other hand, patients with mild kyphosis and multilevel ventral compression can undergo posterior laminectomy and arthrodesis with satisfactory radiographical and clinical results.

For this particular case, we would perform a C4-7 laminectomy and fusion using lateral mass screws. The laminectomy will allow the spinal cord to drift away from the ventral disc bulges, and the fusion will significantly reduce the chance of worsening kyphosis and recurrent spinal cord compression. Some kyphosis correction can be achieved intraoperatively by slightly extending the patient’s neck after the laminectomy has been performed, but before placement of the rods. I will usually ask my assistant to scrub out of the case, and then perform the maneuver while I directly observe the spinal cord. Gentle manipulation combined with intraoperative monitoring allows this to be performed with very little risk. If desired, additional lordosis can be obtained via compression of the instrumentation.
Satisfactory spinal alignment is confirmed intraoperatively by using plain radiography or fluoroscopy.

In patients undergoing multilevel posterior cervical laminectomy and arthrodesis, some surgeons routinely extend their fusion into the upper thoracic spine. This is usually because of concerns regarding the strength of C7 lateral mass screws at the end of the construct, as well as the development of junctional instability by ending the fusion at the cervicothoracic junction. I am not overly concerned about the development of junctional instability in this patient because she is 68 years old, her upper thoracic disc spaces appear to be relatively healthy and wellpreserved, and that her alignment at C7-T1 appears normal without any evidence of listhesis. However, if she did appear to have some significant degeneration or listhesis at this level, then I would not hesitate to cross the junction with the fusion. If  there are questions regarding the suitability of the C7 lateral mass for screw placement, C7 pedicle screws can be used instead.

Revision anterior surgery confers unique challenges. A primary concern is the risk of the surgical approach. The risks
of an anterior cervical approach are well described including graft migration, injury to superior or recurrent laryngeal
nerves, esophageal perforation, persistent dysphagia, dysphonia, odynophagia, dural tear and nonunion. All of these risks increase in revision anterior approaches. Furthermore, the typical benefits of anterior surgery, such as low complication rate, decreased pain and hospital stay, become less pronounced when comparing a revision anterior to a primary posterior surgery. Carreon et al3 evaluated a series of 120 patients with a symptomatic anterior cervical pseudoarthrosis who underwent revision surgery either by an anterior or posterior approach. The authors found that
12 of 27 patients (44%) in the anterior revision group required a second revision surgery for persistent nonunion, whereas only 2 of 93 (2%) of patients in the posterior revision group required a second revision surgery. The authors concluded that posterior fusion is more effective in treating anterior cervical pseudoarthrosis than revision anterior fusion because of its higher fusion rate and lower incidence of repeat revision surgery. Kuhns et al4 found a similarly high rate of fusion in posterior cervical revision approaches.

Dysphagia and recurrent laryngeal nerve dysfunction occur more frequently in myelopathic patients, the elderly, females, multilevel and revision cases. Unfortunately, our illustrative case carries all of these risk factors. The elderly have a higher incidence of preoperative dysphagia as well as a higher incidence of postoperative dysphagia following cervical surgery, regardless of approach. Up to 66% of myelopathy patients have been found to have preoperative swallowing abnormalities with barium swallow study.5 Beutler et al6 evaluated the frequency of recurrent laryngeal nerve (RLN) palsy in a series of 328 patients who underwent anterior cervical spine surgery. The overall rate of RLN palsy was 2.7%, and there was no difference in the occurrence based on side of the approach. However, the palsy
rate was significantly higher, 9.5% in revision cases. An ENT consult with direct laryngoscopy and swallowing evaluation are recommended prior to revision anterior surgery on this patient, adding time and cost. Based on the radiographical and clinical issues associated with this case and mentioned above, a posterior decompression and stabilization would serve this patient.

References
1. Lee TT, Manzano GR, Green BA. Modified open-door cervical expansive laminoplasty for spondylotic myelopathy: operative technique, outcome, and predictors for gait improvement. J Neurosurg. 1997;86:64-68.

2. Suri A, Chabbra RP, Mehta VS, Galkwald S, Pandey RM. Effect of intramedullary signal changes on the surgical outcome of patients with cervical spondylotic myelopathy. Spine J. 2003;3:33-45.

3. Carreon L, Glassman SD, Campbell MJ. Treatment of anterior cervical pseudoarthrosis: posterior fusion versus anterior revision. Spine J. 2006;6:154-156.

4. Kuhns CA, Geck MJ, Wang JC, Delamarter RB. An outcomes analysis of the treatment of cervical pseudarthrosis with posterior fusion. Spine. 2005;30:2424-2429.

5. Frempong-Boadu A, Houten JK, Osborn B, Opulencia J, Kells L, Guida DD, Le Roux PD. Swallowing and speech dysfunction in patients undergoing anterior cervical discectomy and fusion: a prospective, objective preoperative and postoperative assessment. J Spinal Disord Tech. 2002;15:362-368.

6. Beutler WJ, Sweeney CA, Connolly PJ. Recurrent laryngeal nerve injury with anterior cervical spine surgery: Risk with laterality of surgical approach. Spine. 2001;26:1337-1342.

Disclosures

• H Bassewitz: Speaking arrangements: Level B, Stryker. Grand Rounds: Level B, Globus Spine.
• J Kimball: Nothing to disclose.
• L Holly: Consulting: Level D, Medtronic.

 See Results.