Original Article
Endoscopic decompression of cervical spondylotic myelopathy using posterior approach Yad Ram Yadav, Vijay Parihar, Shailendra Ratre, Yatin Kher, Pushp Raj Bhatele1 Department of Neurosurgery, Netaji Subhash Chandra Bose Medical College, 1Department of Radio‑Diagnosis, Madhya Pradesh Magnetic Resonance Imaging Center, Jabalpur, Madhya Pradesh, India
Abstract
Address for correspondence: Dr. Yad Ram Yadav, 105 Nehru Nagar Opposite Medical College, Jabalpur ‑ 482 003, Madhya Pradesh, India. E‑mail: [email protected] Received : 08‑10‑2014 Review completed : 02‑11‑2014 Accepted : 21‑12‑2014
Background: Cervical spondylotic myelopathy, radiculopathy and myeloradiculopathy can be managed by laminoforaminotomy, or bilateral decompression using posterior approach in single or multilevel compression. Posterior endoscopic techniques allow preservation of motion segment and neural decompression without fusion. Materials and Methods: A prospective study of 50 patients of cervical compressive myelopathy with primarily posterior lesion or multilevel anterior compression with acceptable preoperative lordosis was undertaken. Any instability, significant anterior compression, and cervical myelopathy secondary to tumor, trauma, severe ossification of posterior longitudinal ligament, rheumatoid arthritis, pyogenic spondylitis, and destructive spondylo-arthropathy were excluded from the study. There were 5, 23, 12, 10 patients with 2, 3, 4, 5 vertebral body level pathologies, respectively. Results: There were 2, 4, 7, 32, and 5 patients in preoperative Nurick grade 0, 1, 2, 3, and 4, respectively with an average of 2.6 grades. All the patients improved in post‑operative grading with 10, 34, and 6 patients in 0, 1, and 2 grades (average 0.92), respectively. Better outcome was observed in patients with good preoperative grade and in short segment compression on cord. There was no change in cervical Cobb angle after surgery. Follow‑up ranged from 6 to 24 months (averages 19 months). There was small dural tear, minor bleeding from muscles or epidural vessels and temporary C 5 root injury in 1, 3, and 2 patients, respectively. Conclusion: Endoscopic decompression of cervical spondylotic myelopathy is a safe and an effective alternative treatment option in selected patients when pathologic changes are primarily posterior or multi level anterior lesions with acceptable preoperative lordosis. Key words: Cervical pain, cervical spondylosis, cervical vertebrae, intervertebral disc, intervertebral disc degeneration, neck pain, prolapse disc
Introduction Cervical spondylotic myelopathy can produce myelopathy, radiculopathy, and myeloradiculopathy. Access this article online Quick Response Code:
Website: www.neurologyindia.com PMID: *** DOI: 10.4103/0028-3886.149388
640
Spinal cord decompression can be achieved through anterior, posterior, and combined approaches.[1] Posterior approach is often recommended in multilevel posterior cervical compression when pathology is mainly from posterior side.[2] It is possible to treat this problem using a minimally invasive approach.[3] Endoscopic posterior approaches are indicated in lateral and foraminal cervical disc herniations or spurs using laminoforaminotomy[4‑6] and in multilevel compression due to spinal stenosis.[2] Endoscopic decompression obviates the need for muscle dissection and disruption of the posterior tension band Neurology India | Nov-Dec 2014 | Vol 62 | Issue 6
Yadav, et al.: Endoscopic management of cervical spondylotic myelopathy
that can prevent postlaminectomy kyphosis.[1] It does not require sacrifice of a cervical motion segment. Lower incidence of complications, and quicker return to unrestricted activity minimal blood loss, less surgical time, short X‑ray time, and hospital stay[6,7] are the advantages as compared to open techniques. We are reporting our initial experience of endoscopic decompression in 50 patients.
Materials and Methods A prospective study of 50 patients with multilevel cervical compressive myelopathy was undertaken. Detailed history and thorough physical examinations was performed. There were 5, 23, 12, and 10 patients with 2, 3, 4, and 5 vertebral body level pathologies, respectively. Pathologic changes were either primarily posterior or multilevel mild to moderate anterior compression with acceptable preoperative lordosis. Inclusion criteria were strict which resulted in less number of patients recruitment in this group compared to anterior cervical and other posterior cervical approaches. Any instability, significant anterior pathology, cervical myelopathy secondary to tumor, trauma, severe ossification of posterior longitudinal ligament, rheumatoid arthritis, pyogenic spondylitis, and destructive spondylo‑arthropathies were excluded from the study. Cobb angle of segment to be decompressed were measured before and after surgery. All patients underwent preoperative cervical spine X‑rays, and magnetic resonance imaging (MRI) scans [Figure 1]. Decision about level of decompression was based on clinical features and radiological findings; radiological findings sometimes may look more extensive. Clinical features of root involvement, presence of cord changes, and effective canal diameter of less than 10 mm, etc., were used to decide number of level to be decompressed. Post‑operative MRI scan was also performed at 3 months
a
b
c
d
e
f
Figure 1: Preoperative axial (a–d) and sagittal (e and f) magnetic resonance imaging (MRI) scans showing compression from C3 to C6 level
Neurology India | Nov-Dec 2014 | Vol 62 | Issue 6
after surgery or earlier if needed. Post‑operative CT scan was done in some patients. Dynamic X‑rays to detect any instability were also done. Statistical analysis was done using Extended Mantel‑Haenszel chi square test. Surgical procedure: All patients had surgery under general anesthesia. Bilateral decompression using unilateral approach was performed. Foraminotomy was done in selected patients having associated radiculopathy. Patient was turned to prone position and the neck was fixed in slightly flexed position using horseshoe headrest. Fluoroscopic C‑arm was used to exactly localize the site [Figure 2]. The surgeon stood on the side of more compression or on left side if compression was symmetrical. Video monitor was placed opposite to the surgeon. Approximately 2–3 cm skin incision, about 1 cm lateral to midline was made at the spinal level to be decompressed. Subcutaneous tissue, fascia, and muscles were dissected using scissors, insulated monopolar, and bipolar forceps. Para spinal muscles were separated from spinous process. It was difficult to use retractor and two other instruments for dissection because of small incision size. Two limbs of dissecting forceps were used as retractor, cautery or other instruments were used in between the two limbs of forceps. The Destandau microendoscopic set and instruments required for lumbar disc disease were used for this procedure. Destandau set was placed on the lamina and facet joints to be decompressed. Monopolar and bipolar cautery was used to remove any residual muscular and soft tissues overlying the lamina and facet joint and for coagulation. Lamina, facet joint, and interlaminar area were identified. Unilateral laminectomy was done using a high‑speed drill to thin out the lamina. Kerrison rongeurs were used to remove thinned out lamina [Figure 3]. Endoscope was then swung medially to drill the base of spinal process and to undercut opposite lamina [Figure 4]. Ligamentum flavum was left intact until all bony work was complete to prevent dural tear. A small angled
Figure 2: The fluoroscopic image showing localization just before surgery
641
Yadav, et al.: Endoscopic management of cervical spondylotic myelopathy
curette, nerve hook, or Kerrison rongeurs were used to gently separate ligament flavum and underlying dura. Good dural pulsation was observed after removal of the ligamentum flavum. Adjacent cranial or caudal level decompression was performed by vertically pushing endoscopy set cranially or caudally, respectively. Three‑level pathologies can be decompressed using same incision. Another skin incision was added for more than three levels compression [Figure 5]. For example, in case of C3–C7 compression, the skin incisions were made at C4 and C6 levels. It is better to add another incision for more levels of decompression rather than making larger incision or more angulation of endoscope. Smaller incision prevents muscle bleeding by tamponade effect of endoscope tube. We encounter bleeding from muscles in initial learning curve phase because of bigger incision. More angulation of scope allows muscles and other soft tissue to enter inside the endoscopic tube, which can soil telescope tip and adds to the difficulties in the procedure. When two incisions were used, dissection up to the laminas should be done in both operative incisions before removal of laminas otherwise separation of muscle from lamina of later incision, adjoining the decompressed thecal sac, may risk injury to cord. Monopolar or bipolar cautery, bone wax was judiciously used to control bleeding. Absolute hemostasis was achieved before closing. Gently keeping a very small piece of gelfoam between dura matter and adjoining bone could stop epidural bleeding. This gelfoam should be removed before closing the incision. Cauterizing epidural vessel, before removal of epidural soft tissue containing it, prevents epidural bleeding. Bimanual dissection and elevation of neck should be performed to control bleeding. Surgicel or floseal (Baxter) can be used to control bleeding. Most of the bleedings can be controlled by above measures especially after use of floseal. If all these technique fail, inner tube of the Destandau set can be removed and hemostasis can be achieved using microscope keeping outer sheath in place.[8] If nothing works then one might have to use open technique. The endoscope set was removed, and the fascia and skin were closed using standard techniques. Dural tear may occur in severe compression when Kerrison punch is used, and in opposite side decompression. Dural tear may occur when dura matter is caught in the Kerrison punch. Keeping ligamentum flavum until all the bony work is over,[9] using eggshell drilling of bone and removal of thinned out lamina by micro instruments can prevent dural tear. Dural tear due to Kerrison punch can be prevented by partially retracting the Kerrison punch after disconnecting piece of bone to be removed from rest of the lamina and then holding the proximal part of the tissue before it is removed.[9] Diamond burr or
642
a
b
c
d
Figure 3: Endoscopic image showing exposure of (a) lamina after removal of soft tissue, (b) drilling of ipsilateral lamina, (c) removal of partially drilled lamina using Kerrison punch, and (d) decompressed thecal sac at the end of procedure
a
b
c
d
Figure 4: Post‑operative axial computed tomography (CT) scans (a‑d) showing excision of left side laminas with undercutting of contralateral lamina (arrow in B) and partial removal of base of spinous process (arrow in C)
Figure 5: Reconstructed computed tomography (CT) image showing removal of multiple laminas from C3 to C6 on left side
Neurology India | Nov-Dec 2014 | Vol 62 | Issue 6
Yadav, et al.: Endoscopic management of cervical spondylotic myelopathy
ultrasonic bone dissector (remove bone without injuring soft tissue) can be used to decompress under surface of opposite side lamina without injuring dura matter. Small dural tear may need simple placement of Abgel or fat over defect while larger defect may require fascia lata or dural substitute along with tissue glue. Lens soiling during drilling can be minimized by keeping telescope tip as much away as possible from the drill.[10] Continuous suction of fluid helps in decreasing lens soiling. Blood may trickle from skin or surrounding tissue inside the outer tube which may deteriorated image quality or soil lens tip. Proper hemostasis and using proper size sheath prevents this complication. Precision grip should be used to improve precision; if power grip is required precision grip should be added.[8] Insertion of cotton gauges at cranial and caudal ends prevents muscle protrusion inside the sheath. Frequent attention to camera head should be given to prevent its rotation and disorientation during surgery. It is better to keep maximum magnification and adequate illumination for better visualization of structures during surgery.
Results The cohort included 38 males and 12 females and the age ranged between 45 and 75 years. Mean age of male and female patients was 55.42 ± 5.67 (range 46–68 years) and 56.17 ± 9.87 (range 45–75 years), respectively. There were 5, 23, 12, 10 patients with 2, 3, 4, 5 level pathologies, respectively, with an average of 2.54 vertebral level lesions [Table 1]. There were 2, 4, 7, 32, and 5 patients in preoperative Nurick grade 0, 1, 2, 3, and 4, respectively, with an average of 2.6 grade. Post‑operative grading of 0, 1, and 2 was seen in 10, 34, and 6, respectively, with an average of 0.92 Nurick grade (Extended Mantel‑Haenszel Chi square for linear trend χ2 = 55.92; P
Endoscopic decompression of cervical spondylotic myelopathy using posterior approach Yad Ram Yadav, Vijay Parihar, Shailendra Ratre, Yatin Kher, Pushp Raj Bhatele1 Department of Neurosurgery, Netaji Subhash Chandra Bose Medical College, 1Department of Radio‑Diagnosis, Madhya Pradesh Magnetic Resonance Imaging Center, Jabalpur, Madhya Pradesh, India
Abstract
Address for correspondence: Dr. Yad Ram Yadav, 105 Nehru Nagar Opposite Medical College, Jabalpur ‑ 482 003, Madhya Pradesh, India. E‑mail: [email protected] Received : 08‑10‑2014 Review completed : 02‑11‑2014 Accepted : 21‑12‑2014
Background: Cervical spondylotic myelopathy, radiculopathy and myeloradiculopathy can be managed by laminoforaminotomy, or bilateral decompression using posterior approach in single or multilevel compression. Posterior endoscopic techniques allow preservation of motion segment and neural decompression without fusion. Materials and Methods: A prospective study of 50 patients of cervical compressive myelopathy with primarily posterior lesion or multilevel anterior compression with acceptable preoperative lordosis was undertaken. Any instability, significant anterior compression, and cervical myelopathy secondary to tumor, trauma, severe ossification of posterior longitudinal ligament, rheumatoid arthritis, pyogenic spondylitis, and destructive spondylo-arthropathy were excluded from the study. There were 5, 23, 12, 10 patients with 2, 3, 4, 5 vertebral body level pathologies, respectively. Results: There were 2, 4, 7, 32, and 5 patients in preoperative Nurick grade 0, 1, 2, 3, and 4, respectively with an average of 2.6 grades. All the patients improved in post‑operative grading with 10, 34, and 6 patients in 0, 1, and 2 grades (average 0.92), respectively. Better outcome was observed in patients with good preoperative grade and in short segment compression on cord. There was no change in cervical Cobb angle after surgery. Follow‑up ranged from 6 to 24 months (averages 19 months). There was small dural tear, minor bleeding from muscles or epidural vessels and temporary C 5 root injury in 1, 3, and 2 patients, respectively. Conclusion: Endoscopic decompression of cervical spondylotic myelopathy is a safe and an effective alternative treatment option in selected patients when pathologic changes are primarily posterior or multi level anterior lesions with acceptable preoperative lordosis. Key words: Cervical pain, cervical spondylosis, cervical vertebrae, intervertebral disc, intervertebral disc degeneration, neck pain, prolapse disc
Introduction Cervical spondylotic myelopathy can produce myelopathy, radiculopathy, and myeloradiculopathy. Access this article online Quick Response Code:
Website: www.neurologyindia.com PMID: *** DOI: 10.4103/0028-3886.149388
640
Spinal cord decompression can be achieved through anterior, posterior, and combined approaches.[1] Posterior approach is often recommended in multilevel posterior cervical compression when pathology is mainly from posterior side.[2] It is possible to treat this problem using a minimally invasive approach.[3] Endoscopic posterior approaches are indicated in lateral and foraminal cervical disc herniations or spurs using laminoforaminotomy[4‑6] and in multilevel compression due to spinal stenosis.[2] Endoscopic decompression obviates the need for muscle dissection and disruption of the posterior tension band Neurology India | Nov-Dec 2014 | Vol 62 | Issue 6
Yadav, et al.: Endoscopic management of cervical spondylotic myelopathy
that can prevent postlaminectomy kyphosis.[1] It does not require sacrifice of a cervical motion segment. Lower incidence of complications, and quicker return to unrestricted activity minimal blood loss, less surgical time, short X‑ray time, and hospital stay[6,7] are the advantages as compared to open techniques. We are reporting our initial experience of endoscopic decompression in 50 patients.
Materials and Methods A prospective study of 50 patients with multilevel cervical compressive myelopathy was undertaken. Detailed history and thorough physical examinations was performed. There were 5, 23, 12, and 10 patients with 2, 3, 4, and 5 vertebral body level pathologies, respectively. Pathologic changes were either primarily posterior or multilevel mild to moderate anterior compression with acceptable preoperative lordosis. Inclusion criteria were strict which resulted in less number of patients recruitment in this group compared to anterior cervical and other posterior cervical approaches. Any instability, significant anterior pathology, cervical myelopathy secondary to tumor, trauma, severe ossification of posterior longitudinal ligament, rheumatoid arthritis, pyogenic spondylitis, and destructive spondylo‑arthropathies were excluded from the study. Cobb angle of segment to be decompressed were measured before and after surgery. All patients underwent preoperative cervical spine X‑rays, and magnetic resonance imaging (MRI) scans [Figure 1]. Decision about level of decompression was based on clinical features and radiological findings; radiological findings sometimes may look more extensive. Clinical features of root involvement, presence of cord changes, and effective canal diameter of less than 10 mm, etc., were used to decide number of level to be decompressed. Post‑operative MRI scan was also performed at 3 months
a
b
c
d
e
f
Figure 1: Preoperative axial (a–d) and sagittal (e and f) magnetic resonance imaging (MRI) scans showing compression from C3 to C6 level
Neurology India | Nov-Dec 2014 | Vol 62 | Issue 6
after surgery or earlier if needed. Post‑operative CT scan was done in some patients. Dynamic X‑rays to detect any instability were also done. Statistical analysis was done using Extended Mantel‑Haenszel chi square test. Surgical procedure: All patients had surgery under general anesthesia. Bilateral decompression using unilateral approach was performed. Foraminotomy was done in selected patients having associated radiculopathy. Patient was turned to prone position and the neck was fixed in slightly flexed position using horseshoe headrest. Fluoroscopic C‑arm was used to exactly localize the site [Figure 2]. The surgeon stood on the side of more compression or on left side if compression was symmetrical. Video monitor was placed opposite to the surgeon. Approximately 2–3 cm skin incision, about 1 cm lateral to midline was made at the spinal level to be decompressed. Subcutaneous tissue, fascia, and muscles were dissected using scissors, insulated monopolar, and bipolar forceps. Para spinal muscles were separated from spinous process. It was difficult to use retractor and two other instruments for dissection because of small incision size. Two limbs of dissecting forceps were used as retractor, cautery or other instruments were used in between the two limbs of forceps. The Destandau microendoscopic set and instruments required for lumbar disc disease were used for this procedure. Destandau set was placed on the lamina and facet joints to be decompressed. Monopolar and bipolar cautery was used to remove any residual muscular and soft tissues overlying the lamina and facet joint and for coagulation. Lamina, facet joint, and interlaminar area were identified. Unilateral laminectomy was done using a high‑speed drill to thin out the lamina. Kerrison rongeurs were used to remove thinned out lamina [Figure 3]. Endoscope was then swung medially to drill the base of spinal process and to undercut opposite lamina [Figure 4]. Ligamentum flavum was left intact until all bony work was complete to prevent dural tear. A small angled
Figure 2: The fluoroscopic image showing localization just before surgery
641
Yadav, et al.: Endoscopic management of cervical spondylotic myelopathy
curette, nerve hook, or Kerrison rongeurs were used to gently separate ligament flavum and underlying dura. Good dural pulsation was observed after removal of the ligamentum flavum. Adjacent cranial or caudal level decompression was performed by vertically pushing endoscopy set cranially or caudally, respectively. Three‑level pathologies can be decompressed using same incision. Another skin incision was added for more than three levels compression [Figure 5]. For example, in case of C3–C7 compression, the skin incisions were made at C4 and C6 levels. It is better to add another incision for more levels of decompression rather than making larger incision or more angulation of endoscope. Smaller incision prevents muscle bleeding by tamponade effect of endoscope tube. We encounter bleeding from muscles in initial learning curve phase because of bigger incision. More angulation of scope allows muscles and other soft tissue to enter inside the endoscopic tube, which can soil telescope tip and adds to the difficulties in the procedure. When two incisions were used, dissection up to the laminas should be done in both operative incisions before removal of laminas otherwise separation of muscle from lamina of later incision, adjoining the decompressed thecal sac, may risk injury to cord. Monopolar or bipolar cautery, bone wax was judiciously used to control bleeding. Absolute hemostasis was achieved before closing. Gently keeping a very small piece of gelfoam between dura matter and adjoining bone could stop epidural bleeding. This gelfoam should be removed before closing the incision. Cauterizing epidural vessel, before removal of epidural soft tissue containing it, prevents epidural bleeding. Bimanual dissection and elevation of neck should be performed to control bleeding. Surgicel or floseal (Baxter) can be used to control bleeding. Most of the bleedings can be controlled by above measures especially after use of floseal. If all these technique fail, inner tube of the Destandau set can be removed and hemostasis can be achieved using microscope keeping outer sheath in place.[8] If nothing works then one might have to use open technique. The endoscope set was removed, and the fascia and skin were closed using standard techniques. Dural tear may occur in severe compression when Kerrison punch is used, and in opposite side decompression. Dural tear may occur when dura matter is caught in the Kerrison punch. Keeping ligamentum flavum until all the bony work is over,[9] using eggshell drilling of bone and removal of thinned out lamina by micro instruments can prevent dural tear. Dural tear due to Kerrison punch can be prevented by partially retracting the Kerrison punch after disconnecting piece of bone to be removed from rest of the lamina and then holding the proximal part of the tissue before it is removed.[9] Diamond burr or
642
a
b
c
d
Figure 3: Endoscopic image showing exposure of (a) lamina after removal of soft tissue, (b) drilling of ipsilateral lamina, (c) removal of partially drilled lamina using Kerrison punch, and (d) decompressed thecal sac at the end of procedure
a
b
c
d
Figure 4: Post‑operative axial computed tomography (CT) scans (a‑d) showing excision of left side laminas with undercutting of contralateral lamina (arrow in B) and partial removal of base of spinous process (arrow in C)
Figure 5: Reconstructed computed tomography (CT) image showing removal of multiple laminas from C3 to C6 on left side
Neurology India | Nov-Dec 2014 | Vol 62 | Issue 6
Yadav, et al.: Endoscopic management of cervical spondylotic myelopathy
ultrasonic bone dissector (remove bone without injuring soft tissue) can be used to decompress under surface of opposite side lamina without injuring dura matter. Small dural tear may need simple placement of Abgel or fat over defect while larger defect may require fascia lata or dural substitute along with tissue glue. Lens soiling during drilling can be minimized by keeping telescope tip as much away as possible from the drill.[10] Continuous suction of fluid helps in decreasing lens soiling. Blood may trickle from skin or surrounding tissue inside the outer tube which may deteriorated image quality or soil lens tip. Proper hemostasis and using proper size sheath prevents this complication. Precision grip should be used to improve precision; if power grip is required precision grip should be added.[8] Insertion of cotton gauges at cranial and caudal ends prevents muscle protrusion inside the sheath. Frequent attention to camera head should be given to prevent its rotation and disorientation during surgery. It is better to keep maximum magnification and adequate illumination for better visualization of structures during surgery.
Results The cohort included 38 males and 12 females and the age ranged between 45 and 75 years. Mean age of male and female patients was 55.42 ± 5.67 (range 46–68 years) and 56.17 ± 9.87 (range 45–75 years), respectively. There were 5, 23, 12, 10 patients with 2, 3, 4, 5 level pathologies, respectively, with an average of 2.54 vertebral level lesions [Table 1]. There were 2, 4, 7, 32, and 5 patients in preoperative Nurick grade 0, 1, 2, 3, and 4, respectively, with an average of 2.6 grade. Post‑operative grading of 0, 1, and 2 was seen in 10, 34, and 6, respectively, with an average of 0.92 Nurick grade (Extended Mantel‑Haenszel Chi square for linear trend χ2 = 55.92; P
