Correspondence Address: Dr. Bidre N Upendra Vitus Spine Care and Research, Bhagwan Mahaveer Jain Hospital, Vasanth Nagar, Bengaluru - 560 052, Karnataka India
Source of Support: None, Conflict of Interest: None
Background: The success and popularity of the transforaminal approach in the lumbar spine has been made possible by the routine use of pedicle screws in the lumbar spine. Transforaminal approach in the cervical spine can give access to the disc and the vertebral body anteriorly and avoid an additional anterior approach in certain clinical situations. We report technical aspects of transforaminal approach in the lower cervical spine with the authors learning experience. Materials and Methods: Fifteen patients underwent transforaminal approach with cervical pedicle screw (CPS) instrumentation at our institute from July 2011 to October 2014. Five patients underwent foraminal decompression alone (Group-1); 9 patients underwent transforaminal cervical interbody fusion (TCIF) with foraminal decompression, discectomy, and interbody bone grafting (Group-2); and 1 patient underwent partial corpectomy (Group-3). All patients were evaluated for the placement of pedicle screws, for clinical improvement using modified Japanese Orthopaedic Association (mJOA) scoring and interbody graft positioning. The average follow-up was 34.6 months (22–64 months). Results: The average age was 45 years (25–80 years). The average blood loss was 198 ml (100–450 ml) and the average operative time was 142 min (90–200 min). Interbody graft pieces extruded anteriorly in 4 patients (Group-II). The preoperative average mJOA score of 11.4 (0–15) improved to 15.73 (0–18) at final followup. Conclusion: Transforaminal approach in lower cervical spine, though has a learning curve, seems to be a feasible technique along with the use of cervical pedicle screws. Safety and reproducibility of the approach needs to be substantiated with a larger study. Further, TCIF can avoid an additional anterior surgery in certain situations in the cervical spine.
How to cite this article: Mahesh BH, Upendra BN, Raghavendra R, Vijay S, Arun K, Srinivasa R. Transforaminal approach to cervical spine with use of cervical pedicle screws: Technical description of a novel approach. Indian Spine J 2018;1:51-60
How to cite this URL: Mahesh BH, Upendra BN, Raghavendra R, Vijay S, Arun K, Srinivasa R. Transforaminal approach to cervical spine with use of cervical pedicle screws: Technical description of a novel approach. Indian Spine J [serial online] 2018 [cited 2021 Jun 16];1:51-60. Available from: https://www.isjonline.com/text.asp?2018/1/1/51/223441
Transforaminal approach in the lumbar spine has been one of the most popular approaches to access the disc and interbody region since its description by Harms and Rolinger. The success and popularity of the transforaminal approach in the lumbar spine has been made possible by the safe and routine use of pedicle screws in the lumbar spine. Transforaminal approach to the disc and cervical body is not practiced due to the relatively rare use of cervical pedicle screws (CPS) in the lower cervical spine. This is largely due to the feared complications of CPS in the lower cervical spine. Few studies on cadavers have pointed out the high incidence of pedicle screw misplacement in the cervical spine (16.8%–87.5%) and have questioned the clinical use of the same.,, However, clinical studies have shown minimal screw-related complications ,, and have highlighted the advantages of using CPS. A safe and reliable use of CPS in lower cervical spine facilitates the transforaminal approach, similar to that in the lumbar spine. Transforaminal approach can give access to the disc and the vertebral body anteriorly and avoid an additional anterior approach in certain clinical situations. The authors have been routinely using CPS with medial cortical pedicle screw technique ,, for the stabilization of lower cervical spine and wish to share their experience on the feasibility and safety of transforaminal approach in the lower cervical spine.
This is a retrospective study. 15 patients who underwent transforaminal approach with CPS instrumentation at our institute from July 2011 to October 2014 were enrolled. During this period, a total of 45 patients underwent posterior cervical surgeries with CPS instrumentation. Twenty-eight patients had cervical traumatic fractures, 16 patients had cervical spondylotic myelopathy and 1 patient had intradural dumb-bell tumor of the cervical spine. Among these, 15 patients underwent transforaminal approach with partial facetectomy for either transforaminal decompression or discectomy with interbody fusions depending on the requirement of decompression of the cervical spinal cord. Of the 15 patients, 5 patients underwent foraminal decompression alone (Group-1 two patients had cervical traumatic subluxation, one had an intradural tumor and 2 had foraminal stenosis with cervical myelopathy). With more experience, 9 patients underwent transforaminal cervical interbody fusion (TCIF) with foraminal decompression, discectomy, and insertion of bone chips in the interbody region along with pedicle screw-rod instrumentation (Group-2: 7 patients had traumatic subluxation; two had trauma on spondylotic myelopathy;), and 1 patient underwent partial corpectomy (Group-3: 1 patient with C5 burst fracture) with pedicle screw instrumentation [Table 1].
Table 1: Gives details of patients with diagnosis, surgery performed, blood loss, operative time, and follow up (patient 2 expired at 5 weeks after surgery)
The authors used the medial cortical pedicle screw technique for cervical pedicle screw insertion, described earlier., The medial 2/3rd of the lateral mass, forming the roof of the foraminal area, was sequentially drilled with a 3 mm cutting high-speed burr, taking care to stop as the anterior cortex of the superior articular facet was reached. The thinned-out portion of the superior articular facet was taken out with 1–2-mm rongeurs having thin foot plates, over the corresponding cervical root [Figure 1] and [Video 1], available online].
Figure 1: The partial drilling of the lateral mass over the cervical root just above the corresponding pedicle (note that the pedicle has not been instrumented to allow access to the disc between the superior wall of the pedicle and the nerve root) (Reproduced with permission from Mahesh et al.Asian Spine J 2016;10:1007.17)
Adjacent portions of the corresponding laminae were partially removed to expose the dura and the origin of the cervical root. Extensile measures with removal of the adjacent complete laminae and/or total facetectomy were performed depending on the necessity for decompression and the extent of window required [Figure 1] and [Figure 2].
Figure 2: A three-dimensional surface reconstruction of cervical computed tomography scan image with angiography. The lamina and lateral mass are digitally subtracted to visualize the anatomy of the foraminal region. The entry into the interbody region is just above the superior wall of lower pedicle. The UP lies in the trajectory (arrow) of our approach to the disc space and needs to be drilled, preserving its lateral wall. ICA: Internal carotid artery, VA: Vertebral artery, P: Pedicle in cross-section, UP: Uncinate process
The foraminal region consists of the cervical nerve root passing in the lower aspect, closely abutting the cranial wall of the pedicle below and directly overlying the disc space and the uncinate process of the lower vertebra [Figure 2] and [Figure 3].
Figure 3: Schematic superimposition of the spinal cord and the nerve root over Figure 2 to have a clear orientation of the anatomical structures encountered in the transforaminal approach. ICA: Internal carotid artery, VA: Vertebral artery, P: Pedicle in cross-section, UP: Uncinate process
The disc space and uncinate process were accessed by retracting the nerve root cranially. Epidural bleeding was controlled at this stage with a combination of bipolar coagulation, gel foams, and cotton patties. The authors preferred to partially drill the uncinate process with a 2-mm diamond-tipped burr, directed medially, preserving the lateral wall of the uncinate intact, [Figure 2] and [Figure 3] which protected the vertebral artery [Video 2, available online]. A Watson–Cheyne dissector was used to retract the root cranially and rested on the uncinate process acting as a restraint, preventing the breaching of the lateral wall of uncinate process. This enabled us to have adequate access to the disc space, at a more-steeper medial angulation, just above the superior pedicle wall. The entry into the disc at a more-steeper angle also gave us the access to decompress more medial aspect of the central canal, with minimal retraction of the dura [Figure 4] and [Video 2, available online].
Figure 4: Intraoperative picture of patient 12 showing the access to the C4-C5 interbody region at the axilla of the root, just above the superior wall of lower pedicle. P: Pedicle not instrumented, UP: Uncinate process
The access to the foraminal area and the disc was further increased by distracting the interbody region over the rod inserted on the contralateral pedicle screws, similar to lumbar TLIF. The screw in the lower pedicle on the side of interbody approach was deliberately not inserted until the interbody work was completed [Figure 4]. The disc space was entered with a 1-mm micro-disc punch, and the end plates were curetted with small 1-mm curettes [Video 3, available online]. Small chips of bone derived from the lamina were inserted into the interbody region to aide in fusion [Video 4, available online]. The lower pedicle screw on the side of the TCIF was inserted, and the distraction on the contralateral side was reversed before closure.
The average follow-up was 34.6 months (22–64 months). All patients were evaluated for placement of pedicle screws and the interbody graft positioning (Group II) with postoperative CT scans. Pedicle screw perforations were graded with following Criteria: Grade I perforations having ≤50% of the screw outside the pedicle and Grade II perforations having >50% of the screw outside the pedicle. Clinical complications directly related to cervical pedicle screw placement screw were also recorded. All patients were evaluated preoperatively and postoperatively for clinical improvement using modified Japanese Orthopaedic Association (mJOA) scoring.
The average age was 45 years (25–80 years). The average blood loss was 198 ml (100–450 ml) and the average operative time was 142 min (90–200 min). A total of 63 pedicle screws were placed with two patients undergoing only a transforaminal decompression without any pedicle screw instrumentation [patients 4 and 5, [Table 2].
Table 2: Enumerates total number of pedicle screws inserted with grades of perforations
Eighteen screws were misplaced (28.57%), with 10 (15.8%) of them having Grade-I medial perforations, 4 (6.34%) having Grade-II medial perforations, and another 4 (6.34%) having Grade-I lateral perforation. There were no Grade-II lateral perforations. As the authors used the medial cortical pedicle screw technique for placement of pedicle screws,, the lateral perforations were minimal and most of the perforations were Grade-I medial perforations (15.8%).
Interbody graft pieces extruded anteriorly in 4 of the traumatic subluxation patients (Group-II) without any clinical consequences. The preoperative mJOA score of 11.4 (0–15) improved to 15.73 (0–18) at the final follow-up [Table 3]. Patient 2 had complete quadriplegia preoperatively and succumbed to respiratory failure at 5 weeks without any neurological or functional improvement.
Table 3: Enumerates the pre-and postoperative modified Japanese Orthopedic Association scores at final follow-up
Apart from one patient succumbing to respiratory failure, we had one postoperative C5 palsy (patient 14) which improved to normal power, 14 weeks after surgery. We had two patients with dural tear at the axilla of the nerve root due to retraction, which did not affect their clinical outcome.
We wish to share a few patients' details and pictures which helped the authors to move up the learning ladder.
Group-I (patient 3)
A 27-year-old female was diagnosed to have dumb-bell tumor of her left C6 root with intradural extension. She presented with clinically progressive left elbow and wrist weakness with myelopathy signs. She underwent surgery with tumor excision and C5 to C7 pedicle screw-rod instrumentation with left transforaminal approach [Figure 5], [Figure 6], [Figure 7]. She recovered well and has completed 26 months of follow-up with no further complaints. The histopathology confirmed the tumor as Schwannoma. Her preoperative mJOA of 13 had improved to 18 at last follow-up.
Figure 5: Preoperative magnetic resonance imaging of patient 3 (Group-I) showing a dumb-bell tumor of C6 root with intradural extension. The left vertebral artery was pushed anteriorly by the tumor, but was not engulfing it
Figure 6: Intraoperative images of patient 3 (Group-I) showing C5-C6 facetectomy with transforaminal approach for tumor exposure (a), followed by durotomy, tumor dissection (b) and complete excision of tumor ([c] excised tumor in inset) and closure with dural patch (d)
Figure 7: Postoperative images of patient 3 (Group-I) showing surface three-dimensional reconstruction with the left-sided C5-C6 facetectomy with transforaminal approach for tumor removal along with axial computed tomography scans on the right
A 35-year-old male came to us with C5–C6 fracture and bifacetal subluxation with myelopathy (mJOA 13) and left elbow weakness. He had an acute disc herniation on MRI and was counseled about the possible anterior surgery requirement after posterior stabilization. After posterior pedicle screw insertion and reduction with distraction, a left-sided partial facetectomy was done along with laminectomy. The extruded disc was easily removed after retraction of the nerve root cranially. The disc space was accessed by drilling the uncinate process of C6 and a few bone chips were inserted. Although there was anterior extrusion of the graft, C5–C6 fused with slight kyphosis [Figure 8] and [Figure 9]. The patient improved to mJOA 18 at 3-month postoperatively.
Figure 8: Preoperative magnetic resonance imaging and radiographs of patient 9 showing C5-C6 fracture with bifacetal subluxation and acute disc herniation, more toward left foraminal region
Figure 9: Postoperative radiographs of patient 9 showing C5-C6 reduction and pedicle screw instrumentation. Note the extrusion of the bone chips anteriorly in the postoperative radiograph. The patient had uneventful fusion at 1-year followup with kyphosis. The postoperative computed tomography scans show placement of pedicle screws
A 48-year-old male came to us with acute quadriparesis with a history of fall (mJOA 5). The patient had chronic myelopathy symptoms with progressive imbalance for the past 6 months and came to us with acute central cord syndrome due to whiplash, presenting with 1/5 power in upper limbs and 3/5 power in his lower limb. His images showed C4–C5 myelomalacia with posterior ligamentous complex (PLC) injury (hyperintense PLC on T2 images) and moderate cervical kyphosis from C4–C6. Intraoperatively, the authors observed an undisplaced facetal fracture at C5–C6 level [Figure 10]. He underwent C4, C5, and C6 laminectomy and pedicle screw-rod instrumentation from C4 to C7 with decompression and kyphosis correction. The patient had left C4-C5 foraminal disc extrusion for which a transforaminal discectomy and interbody bone grafting was done [Figure 10], [Figure 11], [Figure 12]. The patient developed postoperative infection with meningitis for which he required prolonged ICU care, but improved over 1 year, and at present, his mJOA has improved from 5 to 13.
Figure 10: Preoperative magnetic resonance imaging scans of Patient 12 showing acute whiplash injury over a stenotic C4-C5 segment with C5-C6 undisplaced facetal fracture (black-arrow on right parasagittal magnetic resonance film). Note the myelomalacia changes at C4-5 level with left foraminal disc extrusion at the same level (white-arrow)
Figure 11: Intraoperative pictures of patient 12 showing the C4-C5 left-sided foraminal disc (a) which was removed with root decompression through transforaminal approach. C4-C5 interbody fusion was done with bone grafting after endplate preparation (b and c). The postoperative radiographs and computed tomography scans show the screw placements and C4-C5 interbody bone grafting (d and e)
Figure 12: Radiographs and computed tomography scans of patient 12- at 14-month followup showing C4-C5 interbody bone chips without endplate bridging. However, no graft extrusion was seen anteriorly in this patient as seen in patient 9
A 34-year-old male come to us with the right upper limb weakness (⅖ power) and both lower limb paresis (⅗ power) with spasticity (mJOA-7). The MR Imaging showed C5 burst fracture (C2–C3 fused mass) with posterior displacement of the right-sided bony fragment causing compression of the cord. The PLC was disrupted along with laminar fractures and facet fractures at C4, C5, and C6. The patient also had a vertical split fracture of the C6 body [Figure 13], postoperative AP radiograph]. The patient underwent posterior pedicle screw instrumentation from C4 to C7 with removal of fractured laminae of C3, C4, and C5. The posteriorly displaced, right-sided bony fragment was approached by partial drilling of the medial wall of the C5 pedicle, and we were able to partially drill and push back the fragment anteriorly through this window [Figure 13], [Figure 14], [Figure 15]. The patient did well postoperatively and was able to walk without support at 3-month follow-up. The right upper limb weakness had partially recovered at the final follow-up (mJOA of 7 had improved to 14).
Figure 13: Preoperative magnetic resonance imaging (a,b) of patient 15 showing the posteriorly displaced C5 bony fragment (C2, C3 block vertebra) with disruption of posterior elements. The intraoperative pictures on the right (c,d) show the approach to the fragment through the partially drilled right-sided C5 pedicle (P)
Figure 14: Postoperative images of patient 15 showing the repositioning of the displaced C5 bony fragment. Note the transpedicular approach to the fragment by drilling of the medial wall of the C5 pedicle, the vertical split fracture of the C6 body and the placement of pedicle screws
The present study describes the authors' experience on the relative feasibility of posterior transforaminal approach in the lower cervical spine. Posterior approach to the foramen is not something new. Posterior foraminotomy and discectomy has long been successfully practiced with good results. The approach was limited by the amount of lateral mass and lamina that could be removed due to concerns of inducing instability. The advent of safe placement of CPS has given us the opportunity to enlarge this window laterally with partial or total facetectomy. The transforaminal approach to the disc is from a more lateral zone with partial facetectomy (half to 2/3rd) and at a more-steeper angle than the conventional posterior foraminotomy approach. The approach involves entry into the disc space by drilling the uncinate process with protection of the vertebral artery, placed lateral to the uncinate process. To begin with, the authors were concerned about the proximity and dangers of vertebral artery injury with a more lateral entry into the transforaminal area. However, we found the uncinate process to be a reliable and safe landmark in the foraminal region to avoid injury to the vertebral artery. There have been few reports in the literature of transforaminal approach in patients with lower cervical fracture dislocations with herniated disc. Park et al. reported on removal of the herniated disc fragment through foraminotomy in 7 patients with cervical fracture dislocation and traumatic disc herniation along with use of CPS in a single posterior approach. Nakashima et al. reported on 40 patients with cervical facet dislocations and disc herniation, who underwent posterior pedicle screw instrumentation and reduction. The authors had planned for an anterior procedure in the event of increase in disc herniation during posterior reduction, but they did not require the anterior procedure in any of their patients. To the best of our knowledge, transforaminal approach to the interbody area with endplate curettage and bone grafting has not been described previously.
Anatomical obstacles and limitations of the approach
The major obstacle in this approach is the relatively narrow safe zone available for entry into the interbody region. Similar to the transforaminal approach in the lumbar spine, the safe zone is bound by the superior wall of the pedicle below, the existing root which is retracted cranially, the dura medially, the vertebral artery laterally, and the uncinate process making the floor [Figure 2] and [Figure 3]. The authors used 1–2 mm curettes and 1 mm disc punches to get into the interbody region. However, as can be seen in the computed tomography scans of patient 12 [Figure 12], the endplate preparation was not good enough, and therefore, we could not find bridging bone across the endplates even at 1-year follow up. Further, the bone chips were not good enough to provide a structural interbody support similar to a cage, and we did experience progressive kyphosis before fusion in many of these patients [Figure 9]. The other technical difficulty we faced was of the graft slippage anteriorly in patients with fracture dislocations, presumably due to ruptured anterior longitudinal ligament, and anterior disc annulus. The authors are working toward these discrepancies, and as the technique evolves, we believe that suitable solutions will emerge. The transforaminal approach in the lower cervical spine can be as versatile as in the lumbar spine with routine use of CPS. This study shows that this approach can prevent an additional anterior procedure in certain clinical situations described below:
A traumatic unreduced cervical subluxation with disc herniation: The situation requires an anterior discectomy to prevent further disc herniation during reduction, followed by posterior reduction of facetal dislocation and instrumentation. A single posterior approach with transforaminal discectomy after partial or complete facetectomy, followed by reduction using CPS instrumentation can avoid an additional anterior procedure
In a patient with multilevel kyphotic cervical stenosis with myelopathy, having one or two level foraminal or posterolateral disc herniations: The disc herniations can be approached by a transforaminal approach along with laminectomy for decompression and CPS instrumentation for kyphosis correction, all by a single posterior approach
Burst fractures with posteriorly displaced fractured fragments causing thecal compression with PLC injury or facet fractures: Rigid 3-column stabilization using CPS instrumentation along with transforaminal/transpedicular anterior decompression of the fractured fragment can be done by this approach, similar to thoracolumbar burst fracture surgeries with posterolateral approach to the anterior column
Cervical tumors extending from spinal canal to facets and/or lateral aspect of vertebral body: These can be addressed with single stage posterior approach with pedicle screw-rod instrumentation with complete facetectomy providing access for lateral dissection of the tumor mass from vertebral vessels and also to gain posterolateral access to vertebral body for additional debridement/excision of the tumor.
Advantages of the approach:
The advantage of the transforaminal approach lies in the access gained to the anterior structures of the cervical spine from the posterior aspect. Further, the access to the disc and the vertebral body is from a more lateral zone and at a more-steeper angle compared to the conventional posterior foraminotomy approach. This gives the surgeon access to the lateral and central portions of the interbody area without undue retraction of the cervical cord. This approach can avoid an additional anterior procedure as the goals of anterior decompression rigid 3-column stabilization (using CPS) and interbody fusion can be achieved by a single posterior approach at least in clinical situations described above. However, the technique is still not mature in the sense that it does not provide a rigid interbody spacer support and the interbody fusions occur with segmental kyphosis. A front back procedure with a good interbody spacer and a posterior lateral mass stabilization can give a better result in terms of preserving the cervical sagittal alignment with rigid fusion. Further, using the currently available interbody curette, results in an inadequate interbody endplate preparation, leading to inadequate bridging bone formation. However, these drawbacks can be addressed in the future with appropriate interbody devices and improvization of instrumentation for interbody work in the area. Lateral mass posterior instrumentation in the lower cervical spine is a time-tested familiar method of stabilization and sufficient in most of the situations requiring cervical stabilization. However, the authors believe that with the routine and safe use of cervical pedicle screw instrumentation the evolution of a versatile technique like the transforaminal decompressions and interbody fusions is only a matter of time.
The study is limited by the small size of the study group with heterogeneous pathology and being limited to a single center. However, the present study opens the door to a new approach to the anterior structures in the lower cervical spine from behind. The authors encourage the procedure to be evaluated by surgeons routinely performing cervical pedicle screw insertion in the lower cervical spine for its safety and feasibility.
Transforaminal approach in lower cervical spine, though has a learning curve, seems to be a feasible technique along with the use of CPS and opens the door to a new approach to the anterior structures in the lower cervical spine. Safety and reproducibility of the approach needs to be substantiated with a larger study. Further, TCIF can avoid an additional anterior surgery in certain situations in the cervical spine.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Harms J, Rolinger H. A one-stager procedure in operative treatment of spondylolistheses: Dorsal traction-reposition and anterior fusion (author's transl). Z Orthop Ihre Grenzgeb 1982;120:343-7. [PUBMED]
Ludwig SC, Kramer DL, Balderston RA, Vaccaro AR, Foley KF, Albert TJ, et al. Placement of pedicle screws in the human cadaveric cervical spine: Comparative accuracy of three techniques. Spine (Phila Pa 1976) 2000;25:1655-67.
Park JH, Jeon SR, Roh SW, Kim JH, Rhim SC. The safety and accuracy of freehand pedicle screw placement in the subaxial cervical spine: A series of 45 consecutive patients. Spine (Phila Pa 1976) 2014;39:280-5.
Mahesh B, Upendra B, Vijay S, Arun K, Srinivasa R. Perforations and angulations of 324 cervical medial cortical pedicle screws: A possible guide to avoid lateral perforations with use of pedicle screws in lower cervical spine. Spine J 2017;17:457-65.
Nakashima H, Yukawa Y, Imagama S, Kanemura T, Kamiya M, Yanase M, et al. Complications of cervical pedicle screw fixation for nontraumatic lesions: A multicenter study of 84 patients. J Neurosurg Spine 2012;16:238-47.
Kopjar B, Tetreault L, Kalsi-Ryan S, Fehlings M. Psychometric properties of the modified Japanese Orthopaedic Association scale in patients with cervical spondylotic myelopathy. Spine (Phila Pa 1976) 2015;40:E23-8.