Cervical spondylotic myelopathy and radiculopathy treated by oblique

0 downloads 0 Views 211KB Size Report
Jul 28, 2004 - Oblique corpectomy. Introduction. Cervical spondylotic myelopathy (CSM) and cervical ... spine is approached from the more symptomatic side.
Neurosurg Rev (2004) 27: 252–258 DOI 10.1007/s10143-004-0342-9

ORIGINA L ARTI CLE

R. Kemal Koç . Ahmet Menkü . Hidayet Akdemir . Bülent Tucer . Ali Kurtsoy . İ. Suat Öktem

Cervical spondylotic myelopathy and radiculopathy treated by oblique corpectomies without fusion Received: 29 January 2004 / Revised: 1 April 2004 / Accepted: 5 April 2004 / Published online: 28 July 2004 # Springer-Verlag 2004

Abstract Oblique corpectomy (OC) is an alternative technique for the resection of spondylotic spurs ventral to the cervical spinal cord contributing to cervical spondylotic myelopathy (CSM) and cervical spondylotic radiculopathy (CSR). To evaluate the efficacy of OC for the treatment of cervical spondylotic myeloradiculopathy, we reviewed our experience with OC. Twenty-six patients, 18 males and 8 females, were studied. They averaged 51.3 years of age (range 30–72), Thirteen had myelopathy and 13, radiculopathy. Both magnetic resonance (MR) imaging and computed tomography (CT) were performed preoperatively to define the extent of pathology. The Modified Japanese Orthopedic Association (JOA) score was used to grade the quality of the outcome. Neurologic and radiologic results were assessed. Good and excellent results were observed in 76.9% of the cases with myelopathy. Improvement of radicular symptoms was noted in 84.6% of the cases with radiculopathy. Neuroimaging studies confirmed satisfactory anatomical decompression in all patients. Sagittal alignment decreased from 13° to 12°. The degree of postoperative recovery seemed to be directly related to the age and severity of the preoperative myelopathy. This surgical technique has shown excellent clinical outcomes with fast recovery and adequate anatomical decompression in patients with CSM and CSR. Keywords Anterolateral approach . Cervical spondylotic myelopathy . Cervical spondylotic radiculopathy . Oblique corpectomy

R. K. Koç (*) . A. Menkü . H. Akdemir . B. Tucer . A. Kurtsoy . İ. S. Öktem Department of Neurosurgery, Faculty of Medicine, Erciyes University, 38039 Kayseri, Turkey e-mail: [email protected] Tel.: +90-352-4374574 Fax: +90-352-4372934

Introduction Cervical spondylotic myelopathy (CSM) and cervical spondylotic radiculopathy (CSR) caused by compressive lesions from degenerative spondylotic changes have been surgically treated by various means: subtotal corpectomies with strut grafting, multiple discectomies with or without interbody fusions, anterior foraminotomy, laminectomy, and laminoplasty [2–4, 6, 9–12, 14, 15, 18]. However, the choice of the most appropriate surgical approach for this condition is still controversial. Anterior corpectomy and fusion offers direct resection of pathology but requires bone grafting and fusion with many months of immobilization, while laminectomy or laminoplasty offers indirect posterior decompression without the resection of anterior pathology [2, 3, 7, 16, 18]. An alternative technique, oblique corpectomy (OC), facilitates ventral excision of spondylotic bars while averting the need for fusion [5, 6]. It may be applied for patients with straight or kyphotic axis of the spine. The purpose of this article is to emphasize the main advantages of this technique and to analyze the results of our series.

Patients and methods Patient population A prospective study was performed on 26 patients with CSM (n=13) and CSR (n=13) who had been treated with OC between January 2001 and May 2002. Their ages ranged from 30 to 72 years (mean 51.3 years). Eighteen patients were males and eight were females. Their clinical characteristics are summarized in Table 1. Preoperatively all patients had radiography (lateral neutral, flexion, and extension views), MR imaging and CT scans of the cervical spine. Neurologic recovery, late deterioration, axial pain, radiographic results (alignment, and instability), and surgical complications were determined.

253 Table 1 Summary of clinical data. (CSM cervical spondylotic myelopathy, CSR cervical spondylotic radiculopathy) CSM No. of patients 13 Sex (F/M) 4/9 Age (years) at operation Mean 51.5 Range 30–72 Follow-up (months) Mean 17.3 Range 12–24 No. of operated levels 1 level 3 2 level 8 3 level 1 4 level 1 Operated level C2-3 1 C3-4 2 C4-5 6 C5-6 12 C6-7 5

CSR

Total

13 4/9

26 8/18

52.2 39–65

51.3 30–72

15.0 12–20

16.8 12–24

4 7 0 2

7 15 1 3

0 2 5 11 8

1 4 11 23 13

Surgical technique The detailed surgical technique has been reported by George et al. [5, 6]. The anterolateral aspect of the cervical spine is approached from the more symptomatic side corresponding to the side of the most important compressive element. If either clinically or radiologically there is no predominant side, the approach is made on the side of the smallest vertebral artery (VA). The skin incision was made transverse in one or two levels and longitudinal in three or more levels. A transverse skin incision was made about 3–6 cm long at the anterolateral neck along a skin crease. A longitudinal skin incision was made along the medial edge of the sternocleidomastoid (SCM) muscle for 5–7 cm in the midportion of the lesion. The subcutaneous tissue and the platysma muscle were incised. The natural space between the SCM muscle and the internal jugular vein was widely dissected. There was always a variable amount of fat in the depth of this space. The accessory nerve in this fat was identified when the C2-3 or the C3-4 level needed to be exposed. It was retracted superiorly. The SCM muscle was retracted laterally. The carotid artery, jugular vein, esophagus and trachea were retracted medially, and protected by a wide blade retractor. In this technique, retraction was minimal, because the surgical exposure remained laterally. The sympathetic chain was recognized under the prevertebral aponeurosis. The aponeurosis and sympathetic chain were retracted laterally. The medial portion of the longus colli muscle was excised to expose the transverse processes of the upper and lower vertebrae and the lateral aspects of the vertebral bodies. Before starting the oblique drilling of the vertebral bodies, the VA was identified, because the VA occasionally entered the transverse

foramen at a different level. Careful preoperative assessment of the VA in CT scans will aid in avoiding unexpected findings intraoperatively. The drilling was started on the bodies on both sides of the disc. The drill went into the bodies vertically until the cortical bone of the posterior aspect of the bodies was found. To prevent injury to the VA, the thin cortical bone was left attached to the ligamentous tissue covering the medial portion of the artery. It was fractured at the base of the uncinate process and then removed. The drilling was extended obliquely from the anterolateral corner towards the opposite posterolateral corner. Thus half of the vertebral body was preserved and the anterior longitudinal ligament (ALL) was kept intact. The posterior longitudinal ligament (PLL) was incised and removed with micro-Kerrison rongeurs. In the case of radiculopathy, the intervertebral foramen was fully opened. Neither bone grafting nor osteosynthesis was used. Postoperatively, all patients were asked to wear a cervical collar for 4 weeks, but only when they were standing or walking. Clinical evaluations Preoperative and postoperative neurological function in the patients with myelopathy was evaluated using the modified JOA score [1]. To assess the efficacy of this operation, the recovery rate proposed by Hirabayashi et al. [8] was calculated by the following formula: recovery rate ð%Þ ðpostoperative JOA score  preoperative JOA scoreÞ ð17  preoperative JOA scoreÞ  100: ¼

Postoperative neurological recovery was estimated on the basis of the aforementioned recovery rate, in which 75– 100% was designated excellent, 50–74% good, 25–49% fair, and 0–24% poor.

Neuroradiological evaluation Radiographs were assessed for sagittal alignment by the method outlined by Katsumi et al. [13]. Three categories were used to classify spinal alignment: lordotic, straight, and kyphotic. Fewer than 4° of either kyphotic or lordotic angulation was defined as straight alignment. Sagittal alignment in these postures was obtained by measuring the Cobb angle from the base of C2 to the superior endplate of C7. Spinal instability was quantified by the increased intervertebral angle difference in flexion-extension of more than 12° or displacement of the vertebral body over 3.5 mm in flexion [19]. The degree of spondylosis was documented preoperatively with both MR and CT. Before the patient was discharged, the extent of drilling was evaluated by CT. Spine stability was demonstrated at

254 Table 2 Neurological evaluation and radiological results Preoperative

Postoperative

JOA score Mean 10.1(±4.3) Range 3–16 Cervical curvature index Mean 13° Range 0°, 35°

14.1(±4.1) 6–17 12° −6°, 54°

Among patients with radiculopathy, 84.6% were completely relieved of their preoperative radicular symptoms. A good result with minimal residual symptoms was noted in 7.7%, whereas 7.7% had persistent, although decreased, pain. Fig. 1 Preoperative CT scan showing anterior osteophytic compression

1.5, 3, 6 and 12 months by dynamic radiography (flexionextension), and spinal cord decompression was demonstrated at 6 months by MR imaging. Predictors of outcomes The influence of the patient’s age (younger than or older than 50 years), number of levels (more or less than two levels), symptom duration (more than or less than 1 year), preoperative neurological function, cervical curvature, and presence or absence of preoperative high signal intensity within the spinal cord, as revealed by T1- and T2weighted MR imaging, were investigated. Statistical analysis Analysis of all measurements (mean values and ranges) was performed, and each parameter was calculated. The Spearman’s correlation analysis and Wilcoxon signedranks sum test were used to correlate and differentiate between variables, respectively. The statistical significance difference was set at p50 Duration of symptoms (years) ≤1 >1 No. of levels decompressed 1–2 levels 3–4 levels High-intensity areas on T2-weighted MR

Excellent Good Fair Poor

7 1

0 2

0 3

0 0

8 0

2 0

1 2

0 0

7 1 4

1 1 1

3 0 2

0 0 0

255 Fig. 2a, b Preoperative MR images showing severe spinal cord compression

Surgical complications

Discussion

Surgical complications are summarized in Table 4. One patient with myelopathy due to ossification of PLL had persistent brachialgia postoperatively. She was reoperated on and a good clinical outcome was achieved. One patient with myelopathy developed axial pain 1 month postoperatively. Imaging studies showed kyphotic deformity at the same level. She refused reoperation. Another two patients with axial pain were followed up. Mortality was none and morbidity was limited to a Horner’s syndrome in 30.7% (permanent in 7.7% of those) of the patients, and to a CSF leak in one case, and contralateral side foraminal stenosis developed in one patient. He was reoperated on and anterior fusion was performed.

CSM is a disease of the cervical spinal cord that results from circumferential compression of the degenerative cervical spine or a congenitally narrow spinal canal. Surgical recommendations must be based on patient characteristics, symptoms, and neuroradiologic findings.

Fig. 3 Postoperative CT scan showing oblique corpectomy from right side

Surgical approaches and outcomes Standard surgical treatment for CSM has been either direct anterior excision, necessitating fusion, or indirect posterior decompression. Posterior approach Different techniques of posterior decompression have been reported, such as extensive or limited laminectomy, posterior foraminotomy, and, more recently, open door laminoplasty [2, 12, 18]. Although the posterior procedure preserves the anterior motion segments and does not usually require bone-graft fusion, compressive spondylotic spurs are not usually resected. Laminectomy historically yields poor results from late deformity and late neurologic deterioration. Laminoplasty was developed to address cervical stenosis of three or more segments and compares favorably with anterior corpectomy and fusion for neurologic recovery. Laminoplasty had a lower complication rate than corpectomy and strut grafting, but postoperative axial pain was observed in 40% of cases, kyphotic deformity developed in 6%, and the range of motion was reduced severely [18]. Anterior approach Anterior decompression has now become more widely used. It can be achieved using different techniques; anterior discectomy with or without

256 Fig. 4a, b Postoperative MR images showing good decompression of spinal cord

Table 4 Complications

Type of complications

No.

Treatment

Persistent brachialgia Kyphosis/axial pain Axial pain Horner’s syndrome CSF fistula Contralateral foraminal stenosis

1 1 2 8 1 1

Reoperations with good outcome Kyphosis demonstrated/surgery refused Follow-up Permanent 7.7% Lumbar punctions with good outcome Reoperation anteriorly

fusion, anterior subtotal corpectomy, and anterolateral decompression. Anterior cervical discectomy with fusion is an excellent option for one- or two-level spondylosis. Anterior corpectomy may provide an improved decompression and is ideal for patients with kyphotic deformity [18]. Anterior corpectomy technique requires bone fusion with or without spinal instrumentation, but degenerative changes at adjacent vertebral levels frequently result in long-term morbidity. Wada et al. [18] demonstrated that listhesis exceeding 2 mm developed at 38% of the upper adjacent levels, osteophyte formation occurred at 54% of the lower adjacent levels, and axial pain was observed in 15%. The mean vertebral range of motion had decreased from 39.4° to 19.2° in 49% of patients by the final followup assessment. In addition, conventional corpectomy series were focused excessively on graft morbidity and pseudoarthrosis [3, 14, 18]. Technique in oblique corpectomy The OC technique reported here accomplishes direct decompression of the spinal cord just as the anterior corpectomy techniques does, but does not require bone fusion or postoperative immobilization. As with other anterior techniques, this technique is applicable only if the compressive lesion is located anteriorly to the spinal cord.

(3.8%) (3.8%) (7.7%) (30.7%) (3.8%) (3.8%)

The drilling is much more extended laterally on the operative side, and OC totally opens the intervertebral foramen. On the opposite side, the drilling cannot go so far laterally. At best, it reaches the opposite pedicle. Consequently, a bilateral radiculopathy cannot be treated in one stage [4–6]. Such patients are operated by discectomy and fusion with cages or corpectomy and fusion with instrumentation. Multilevel anterior discectomy also seems a less efficient technique than OC. In many cases of multisegmental CSM, the compressing bone disease results not only from osteophytes along the cartilaginous end plate but also from the vertebral body itself [6]. Neuroradiological and clinical assessment Follow-up CT scans and MR imaging of the cervical spine proved that adequate anterior decompression of the spinal cord was achieved. Postoperative change in alignment of the cervical spine is important when assessing the efficacy of this surgical technique. In our series, physiological lordosis decreased from 13° to 12°. Kyphotic deformity developed in one case (4%). Most of the anterior intervertebral segments were preserved with this technique. But, most of these patients had experienced advanced spondylotic changes with limited motion at the

257

involved intervertebral motion segments [11]. In our series, contralateral side foraminal stenosis developed in one patient. The main advantages of this technique have been the elimination of bone fusion and immobilization, thus avoiding bone site and instrumentation complications, and degenerative changes at adjacent vertebral levels. The number of levels has never been considered to be a limitation of this technique [6]. In our cases, the number of levels ranged from one to four. The disadvantage of this technique, however, is that the VA had to be dissected and the potential injury for the VA was present. Surgical time is longer and axial pain occurred rarely. Only patients who had hard and collapsed discs were candidates for this procedure. Patients whose dynamic plain radiographs showed more than 2 mm of movement between two adjacent vertebral bodies were also excluded. Bone resection is limited in OC technique. More than half of the vertebral bodies are preserved. The risk of further spinal instability is smaller after oblique transcorporeal drilling, because only the posterior lateral part of the vertebral body is removed and because the ALL is preserved [6]. Intraoperative CT may enhance the safety of the procedure. In this technique, the risk of instability is mainly related to the resection of the nucleus pulposus of one or several well-hydrated discs [4, 5]. The instability may worsen postoperatively. However, when the disc is already dehydrated and collapsed, as is commonly observed in spondylosis of the cervical spine, no grafting or osteosynthesis is necessary. This has been demonstrated well in cases of CSM treated by decompression with OC [4–6]. This procedure is most appropriate for older patients with more degenerated discs. In our cases, postoperative axial pain was observed in 11.5% of patients (kyphosis with axial pain in one, only axial pain in two).

9% experienced discrete residual Horner’s syndrome. Horner’s syndrome occurred in 30.7% of our patients. Permanent Horner’s syndrome occurred in 7.7% of these. The dissection of the accessory nerve was necessary only to expose the C2-3 and C3-4 levels. The VA control permits the safe drilling of the posterolateral corner of the vertebral bodies and control of the distal nerve root. Predictors of clinical outcomes Patients with multisegmental areas of high signal intensity on T2-weighted MR images tended to have poorer surgical results. However, the transverse area of the spinal cord at the level of maximum compression was a better prognostic indicator [17]. Low-signal intensity changes on T1weighted sequences indicated a poor prognosis [16]. In our patients, there was no relation between results and preoperative high signal intensity at T2-weighted MR imaging. The degree of postoperative recovery in the CSM seemed to be directly related to the age and severity of symptoms (p=0.002, p=0.009, respectively). Indication of oblique corpectomy Indications of the technique of OC without fusion are data similar to those used for other anterior decompression techniques; predominance of anterior osteophytic spurs. If the lordosis of the cervical spine is preserved, a posterior decompression is probably the best choice because the spinal cord may then move posteriorly. If the axis of the cervical spine is straight or kyphotic, it seems more efficient to realize an anterior decompression.

Conclusion Surgical complications The surgical approach lateral to the vasculonervous elements leads directly to the lateral aspect of the vertebral bodies and the transverse processes. There is no vasculonervous element in the way except for the accessory nerve at the C2-3 and C3-4 levels. Therefore, the medial portion of the longus colli muscle had only to be resected. Before dividing this muscle, the sympathetic nerve running under the aponeurosis must be identified and gently retracted laterally. If the sympathic nerve is manipulated, the patient may exhibit Horner’s syndrome postoperatively. However, if the main trunk of the sympathetic nerve is kept intact, no or mild Horner’s syndrome occurs, and recovery is rapid. Because the sympathetic nerve passes along the lateral portion of the longus colli muscle, to prevent the incidence of Horner’s syndrome, the longus colli muscle is not incised laterally to the anterior tubercle of the transverse process. George et al. [6] showed that Horner’s syndrome had occurred in 57% of the patients. In the vast majority, Horner’s syndrome was mild and completely resolved. However,

OC can provide safe and successful removal of the spondylotic spur located anteriorly to the spinal cord. No fusion is required regardless of the number of levels. This surgical technique has shown excellent clinical outcomes with fast recovery and adequate anatomical decompression in patients with CSM and CSR. In our experience, this technique is, therefore, an alternative to the anterior transcorporeal approach.

References 1. Chiles BW III, Leonard MA, Choudhri HF, Cooper PR (1999) Cervical spondylotic myelopathy: patterns of neurological deficit and recovery after anterior cervical decompression. Neurosurgery 44:762–769 2. Fager CA (1973) Results of adequate posterior decompression in the relief of spondylotic cervical myelopathy. J Neurosurg 38:684–692 3. Fessler RG, Steck JC, Giovanini MA (1998) Anterior cervical corpectomy for cervical spondylotic myelopathy. Neurosurgery 43:257–265

258 4. George B, Lot G (1994) Oblique transcorporeal drilling to treat anterior compression of the spinal cord at the cervical level. Minim Invasive Neurosurg 37:48–52 5. George B, Zerah M, Lot G, Hurth M (1993) Oblique transcorporeal approach to anteriorly located lesions in the cervical spinal canal. Acta Neurochir (Wien) 121:187–190 6. George B, Gauthier N, Lot G (1999) Multisegmental cervical spondylotic myelopathy and radiculopathy treated by multilevel oblique corpectomies without fusion. Neurosurgery 44:81–90 7. Herkowitz HN, Kurz LT, Overholt DP (1990) Surgical management of cervical soft disc herniation. A comparison between the anterior and posterior approach. Spine 15:1026– 1030 8. Hirabayashi K, Miyakawa J, Satomi K, Maruyama T, Wakano K (1981) Operative results and postoperative progression of ossification among patients with ossification of cervical posterior longitudinal ligament. Spine 6:354–364 9. Jho HD (1996) Microsurgical anterior cervical foraminotomy for radiculopathy: a new approach to cervical disc herniation. J Neurosurg 84:155–160 10. Jho HD (1997) Decompression via microsurgical anterior foraminotomy for cervical spondylotic myelopathy. Technical note. J Neurosurg 86:297–302 11. Jho HD (1997) Spinal cord decompression via microsurgical anterior foraminotomy for spondylotic cervical myelopathy. Minim Invasive Neurosurg 40:124–129

12. Jho HD, Kim MH, Kim WK (2002) Anterior cervical microforaminotomy for spondylotic cervical myelopathy: part 2. Neurosurgery 51(Suppl 2):54–59 13. Katsumi Y, Honma T, Nakamura T (1989) Analysis of cervical instability resulting from laminectomies for removal of spinal cord tumor. Spine 14:1171–1176 14. Lee TT, Manzano GR, Green BA (1997) Modified open-door cervical expansive laminoplasty for spondylotic myelopathy: operative technique, outcome, and predictors for gait improvement. J Neurosurg 86:64–68 15. Lunsford LD, Bissonette DJ, Zorub DS (1980) Anterior surgery for cervical disc disease. Part 2. Treatment of cervical spondylotic myelopathy in 32 cases. J Neurosurg 53:12–19 16. Morio Y, Teshima R, Nagashima H, Nawata K, Yamasaki D, Nanjo Y (2001) Correlation between operative outcomes of cervical compression myelopathy and MRI of the spinal cord. Spine 26:1238–1245 17. Wada E, Yonenobu K, Suzuki S, Kanazawa A, Ochi T (1999) Can intramedullary signal change on magnetic resonance imaging predict surgical outcome in cervical spondylotic myelopathy? Spine 24:455–461 18. Wada E, Suzuki S, Kanazawa A, Matsuoka T, Miyamoto S, Yonenobu K (2001) Subtotal corpectomy versus laminoplasty for multilevel cervical spondylotic myelopathy: a long-term follow-up study over 10 years. Spine 26:1443–1447 19. White AA III, Johnson RM, Panjabi MM, Southwick WO (1975) Biomechanical analysis of clinical stability in the cervical spine. Clin Orthop 109:85–96