- Email: [email protected]
Stand-alone interbody cage versus anterior cervical plate for treatment of cervical disc herniation: Sequential changes in cage subsidence
Available online at www.sciencedirect.com
Journal of Clinical Neuroscience 15 (2008) 1017–1022 www.elsevier.com/locate/jocn
Clinical Study
Stand-alone interbody cage versus anterior cervical plate for treatment of cervical disc herniation: Sequential changes in cage subsidence Shunsuke Fujibayashi *, Masashi Neo, Takashi Nakamura Department of Orthopaedic Surgery, Graduate School of Medicine, Kyoto University, 54 Shogoin-Kawaharacho, Sakyo-ku, Kyoto 606-8507, Japan Received 10 April 2007; accepted 13 May 2007
Abstract Anterior cervical discectomy and fusion with an autogenous iliac bone graft is the gold standard treatment for cervical disc herniation. However, autologous bone grafts obtained from the anterior iliac crest are associated with significant donor-site morbidity and complications. To decrease bone graft-related problems, several types of interbody fusion cage have been developed and are used widely in clinical practice. We compared the clinical and radiological outcomes for two surgical procedures used to treat cervical disc herniation: the stand-alone interbody cage and autologous iliac bone grafting with an anterior plate. The clinical results did not differ between patients treated with the two procedures. The stand-alone cage was less invasive and had less donor-site morbidity. In patients treated with the bone graft and plate, the alignment of the fused segment was maintained in all but one patient, who exhibited nonunion. In contrast, in the cage-treated group, 44% of patients exhibited loss of lordotic alignment of more than 5° and cage subsidence of 3 mm or more. All cage subsidence occurred within 3 months of surgery. Although the stand-alone cage was a less invasive and more effective procedure to treat cervical disc herniation, surgeons should consider the possible drawbacks of the associated subsidence. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Cervical cage; Interbody fusion; Cervical disc herniation; Subsidence
1. Introduction Anterior cervical discectomy and fusion (ACDF) is the gold standard treatment for cervical disc herniation. Many technical modifications have been reported since its original description by Smith and Robinson1 and the later report by Cloward in 1958.2 Autologous iliac bone grafting is used widely either with or without a supplemental anterior plate. However, autologous bone grafts obtained from the anterior iliac crest are associated with significant donor-site morbidity and complications including severe acute pain, neurological injury, hematoma formation, infection and chronic pain.3,4 In patients treated without the plate, additional complications can include graft extrusion, collapse, and failure of fusion leading to kyphosis and *
Corresponding author. Tel.: +81757513362; fax: +81757518409. E-mail address: [email protected] (S. Fujibayashi).
0967-5868/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2007.05.011
pseudoarthrosis.5 The addition of an anterior plate system reduces the problem of graft extrusion and collapse but is itself associated with problems such as screw or plate dislodgement, dysphagia, and soft-tissue injury.6,7 To decrease the risk of such complications, several types of interbody fusion cage have been developed recently and are used widely in clinical practice. Cages are categorized roughly by shape into cylindrical and rectangular types. A sufficiently distracted intervertebral space after discectomy can be stabilized for multidirectional movement by tension forces of the residual annulus and ligaments. In clinical practice, distraction with restoration of disc height in a degenerative spine increases neuroforaminal volume and contributes to nerve root decompression. In this study, we used a stand-alone rectangular-type titanium cage to treat cervical disc herniation and compared the clinical and radiological results with those obtained using the conventional anterior plating method.
1018
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
2. Materials and methods
2.1. Surgical procedures
Eighteen consecutive patients with single-level cervical disc herniation that was refractory to conservative treatment were treated surgically. Nine patients were treated with the stand-alone cage procedure (cage group) by S.F. and another nine patients were treated with the anterior plating method (plate group) by M.N. In our institute, the treatment options for single-level cervical disc herniation include ACDF, posterior foraminotomy, and expansive laminoplasty. ACDF was selected for patients with myelopathy and/or radiculopathy with cord compression radiologically. Posterior foraminotomy was selected for patients with radiculopathy without cord compression; and expansive laminoplasty was selected for patients with disc herniation combined with symptomatic developmental spinal canal stenosis. One of the 18 patients with failed posterior foraminotomy was successfully treated using ACDF. The demographic and clinical data were similar in the two groups (Table 1). The mean age at the time of surgery was 53.3 years (range, 31–75 years) in the cage group and 49.0 years (28–64 years) in the plate group. The mean follow-up period was 23.1 months (12–30 months) in the cage group and 25.2 months (13–36 months) in the plate group. Preoperatively, 11 patients had myelopathy and seven had radiculopathy. Table 2 summarizes the levels involved. The most common cervical level involved was C5–6. The operative time and intraoperative blood loss were compared to assess the surgical invasiveness for the two methods. The Japanese Orthopaedic Association (JOA) score and recovery rate (Hirabayashi’s method) were compared before surgery, 3 months after surgery, and at the latest follow-up examination for each group. Donor-site pain was assessed at 3 months and the latest follow-up examination.
Surgical procedures were performed using a standard left anterolateral approach. After insertion of a cervical spine distracter, complete discectomy and neural decompression were performed using a surgical microscope. The cartilaginous endplate was removed completely to expose the cortical endplate. The bony endplate was preserved as much as possible to prevent cage subsidence. An appropriate-sized cage (Syncage-C; Synthesis, Paoli, PA, USA) was filled with autologous cancellous bone harvested as a cylinder from the left anterior iliac crest through a mini-incision using a special device. The cage size was determined by both preoperative templating and intraoperative evaluation using a trial cage to confirm initial stability. The cage was inserted into the disc space by using an impactor and cage stability was confirmed after the distracter was removed. Patients remained in a soft collar for 4 weeks postoperatively. Anterior plating was performed according to the Smith–Robinson technique, in which tricortical autologous iliac bone is harvested from the iliac crest and grafted between the vertebral bodies under manual traction. Supplemental anterior plate fixation was applied using an Atlantis plate system (Medtronic Sofamor Danek, Minneapolis, MN, USA).
Table 1 Patient demographic data
Age (years) Male/female Follow-up period (months) Myelopathy/radiculopathy
Cage group
Plate group
Significance
53.3 (31–75) 3/6 23.1 (12–30) 5/4
49.0 (28–64) 6/3 25.2 (13–36) 6/3
NS NS
Statistical analysis was performed using Student’s t-test and significance was set at p < 0:05. NS, not significant.
Table 2 Levels involved Level
No. patients
C3/4 C4/5 C5/6 C6/7
2 4 7 5
2.2. Radiological assessments To assess bony union, three different radiological parameters on lateral dynamic radiographs were determined at several times: before surgery, immediately after surgery, every 3 months after surgery, and at the latest follow-up examination. The fused segment angle (FSA) was defined as the angle formed between the lines drawn parallel to the cranial endplate of the cranial vertebrae and the caudal endplate of the caudal vertebrae. More than 2° motion at flexion–extension was considered to indicate nonunion. The interspinous process distance at flexion– extension was measured between the tips of both spinous processes. More than 2 mm motion at flexion–extension was considered to indicate nonunion. In addition, radiolucency >50% over the anteroposterior distance of the interface between the endplates and implants was defined as nonunion. Multidirectional CT assessment of the coronal and sagittal reconstruction views was also performed 6 and 12 months after surgery. Union was considered to have occurred when bony trabecular orientation was visible (Fig. 1). Radiologically successful fusion was considered to have occurred when all of the three radiological parameters mentioned above and CT assessments indicated fusion. Cage subsidence was calculated from the change of fused segment height (FSH), which was ascertained using the lengths of lines drawn between the center of the cranial endplate of the cranial vertebrae and the center of the caudal end of the caudal vertebrae. A change of 3 mm or more was defined as significant cage subsidence.
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
Fig. 1. Sagittal (left) and coronal (right) reconstructed CT images of the cervical vertebrae of a patient who underwent interbody cage placement for cervical disc herniation. A continuous bony trabecular pattern can be seen through the cage.
Statistical analysis was performed using Student’s t-test; p < 0:05 was accepted as significant. 3. Results The clinical results are summarized in Table 3. Although not significant, the level of surgical invasiveness was lower in the cage group than in the plate group. The surgical time was 124.4 min in the cage group and 149.7 min in the plate group. The intraoperative blood loss was 13.6 mL in the cage group and 19.2 mL in the plate group (not significant, NS). The JOA recovery rate was 96.2% in the cage group and 79.3% in the plate group (NS). Three months after surgery, 22.2% of patients in the plate group complained of donor-site pain compared with none in the cage group. No patient in either group complained of donor-site pain at the final follow-up. We observed no major complications related to surgical interventions in either group throughout the follow-up periods. Radiologically successful fusion was achieved by 88.8% of patients in both groups. The mean time until bony union was 5.6 months (3–18 months) in the cage group and 7.7 months (5–12 months) in the plate group (NS).
Table 3 Clinical and radiological results
Operative time (min) Blood loss (mL) JOA score recovery rate (%) Donor site pain (%) At 3 months At latest follow-up Bony union (%) Time until bony union (months)
Cage group
Plate group
Significance
124.4 (82–181) 13.6 (0–61) 96.2 (73.7–100)
149.7 (130–200) 19.2 (0–63) 79.3 (59.3–100)
NS NS NS
1019
In the cage group, the mean FSA was –1.78° before surgery, 3.89° immediately after surgery, 1.67° 3 months after surgery, and –0.78° at the final follow-up examination. The FSA values differed significantly between before surgery and immediately after surgery (p ¼ 0:0041), between immediately after surgery and 3 months after surgery (p ¼ 0:0304), and between 3 months after surgery and the final follow-up examination (p ¼ 0:0237). The difference in FSA between before surgery and the final follow-up examination was not significant. In the cage group, the mean FSH was 33.11 mm before surgery, 34.67 mm immediately after surgery, 33.33 mm 3 months after surgery, and 32.78 mm at the final followup examination. The FSH values differed significantly between before and immediately after surgery ðp ¼ 0:0278Þ, and between immediately after and 3 months after surgery ðp ¼ 0:0114Þ. The differences in FSA between before surgery and the final follow-up examination, and between 3 months after surgery and the final follow-up examination, were not significant. In contrast, in the plate group, the mean FSA was 3.22° before surgery, 0.22° immediately after surgery, 0.22° 3 months after surgery, and 0.33° at the final follow-up examination. The FSA differed significantly between before and immediately after surgery ðp ¼ 0:0122Þ and between before surgery and the final follow-up examination ðp ¼ 0:0160Þ. In the plate group, the mean FSH was 35.22 mm before surgery, 36.22 mm immediately after surgery, 35.67 mm 3 months after surgery, and 35.22 mm at the final follow-up examination. Only the difference between the values before and immediately after surgery was significant ðp ¼ 0:0085Þ. The acquired FSH gradually decreased to presurgery values, although this was not significant. These results indicate that, although FSH was not maintained in the plate group, the FSA was corrected and maintained through the follow-up period. Overall, the alignment of the fused segment was maintained well in all but one case of nonunion in the plate group. In contrast, a loss of lordotic alignment of >5° occurred in 44% of patients (Fig. 2a,b), and cage subsidence of P3 mm occurred in 44% of cases (Fig. 3a,b). All cage subsidence occurred within 3 months of surgery, but no progressive subsidence was observed after this time. Interestingly, all cages showed about 1–2 mm of minimum subsidence in the early postoperative period. No anterior or posterior migration was observed. Fig. 4 presents the results of a typical successful union and Fig. 5 a typical nonunion with progressive cage subsidence. 4. Discussion
0 0 88.8 5.6 (3–18)
22.2 0 88.8 7.7 (5–12)
NS
Statistical analysis was performed using Student’s t-test and significance was defined at p < 0:05. JOA, Japanese Orthopaedic Association; NS, not significant.
The possible advantages of the stand-alone interbody cage are that it is less invasive; requires a simple surgical procedure, short operating time, and smaller amount of harvested bone; causes less blood loss; is associated with a shorter period of postoperative external immobilization; and can correct cervical kyphosis and restore disc height.
1020
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
Fig. 2. Sequential change in the fused segment angle (FSA) in the cage (a) and plate (b) groups. The loss of alignment was >5° in 44% of patients in the cage group (a), but alignment was maintained well in the plate group (b) throughout the follow-up period. Pre-op, preoperative; Post-op, postoperative; 3 months, 3-month follow-up; Latest F/U, latest follow-up examination.
Fig. 3. Sequential changes in fused segment height (FSH) in the cage (a) and plate (b) groups. Forty-four percent of patients in the cage group (a) showed a decrease in FSH of P3 mm, which occurred mainly within the 3 months following the operation. In contrast, FSH was maintained well in the plate group (b) during the follow-up period. Pre-op, preoperative; Post-op, postoperative; 3 months, 3-month follow-up; Latest F/U, latest follow-up examination.
Fig. 4. Sequential lateral X-rays of the cervical vertebrae of a patient with successful fusion immediately after, 3 months after, and 1 year after surgery for interbody cage placement. Cage subsidence of 1 mm with interlocking with the cage and vertebral endplate were recognized at 3 months and no progression of cage subsidence was evident at the 1 year follow-up. Post-op, postoperative.
Although the present study had some limitations, including a small sample size and no patient randomization, our results showing significant cage subsidence and loss of cervical alignment show that there are potential disadvantages as well as possible advantages. In the cage procedure, correction of operative alignment and restoration of disc height were achieved more effec-
tively and easily than with the conventional approach. However, the acquired alignment and height tended to return gradually toward presurgical values. In contrast, plating maintained the acquired lordotic alignment to achieve successful bony union, although disc height tended to return gradually to the presurgical value. These results could be attributed to surgical procedures, such as the use of a
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
1021
Fig. 5. Sequential lateral X-rays of the cervical vertebrae of a patient with nonunion, immediately after, 3 months after, and 1 year after surgery interbody cage placement. An oversized wedge-shaped cage was inserted with excessive distraction. Cage subsidence of 4 mm was recognized at 3 months, and further cage subsidence and a radiolucent line between the cage and endplate was observed at the 1 year follow-up. Post-op, postoperative.
special distracter and impaction device in the cage group, which increase the disc height during surgery. We found it difficult to maintain the acquired height in the longitudinal direction in either procedure. Although effective disc and foraminal height was restored by surgery, this restoration was lost gradually until the posterior facet joints were stabilized. The stand-alone cage has been used widely in clinical practice, and successful clinical results have been reported in more than one study, although the reliability of this technique remains controversial. In a multicenter study undertaken to obtain US Federal Drug Administration approval, in which the cylindrical cage (BAK/C; Sulzer Spine-Tech Minneapolis, MN, USA) was compared with noninstrumented bone-only fusion, similar success rates were achieved for the two techniques.8 In the multicenter study, the complication rate associated with ACDF was 20%, which was lower than the overall complication rate of 12% for the cage. In a case-series involving 47 patients, 98% of patients treated with the BAK/C cage had achieved solid fusion at an average of 6 months after the operation, but the use of a cervical intervertebral cage in anterior cervical microdiscectomy did not prevent a reduction in the height of the cervical disc space after surgery.9 In another study in which the cage was compared with a tricortical graft it was shown that tricortical graft fusion is cheaper than cage fusion and is more effective in reducing the pain score.10 The shape of the implant and preparation of the endplates both influence the tendency for cage subsidence of an intervertebral implant. A large contact surface together with intact endplates decreases the tendency for subsidence, whereas a small surface area and destructive preparation of the end plates increases the risk of subsidence. In a large comparative clinical study of the BAK/C cage, WING cage (Medinorm, Quierschied, Germany), RABEA cage (Signus, Alzenau, Germany), NOVUS cage (Sofamor Danek, Memphis TN, USA), and CBK cage (Scient’X, Paris, France), asymptomatic anterior cage migration and posterior intracanal migration were associated with cylindrical devices.11 Significant subsidence was observed with cylindrical devices, although minor subsidence of 1–2 mm was observed frequently with the other devices. The authors
concluded that rectangular cages allow for better preservation of sagittal balance or correction because of a lesser tendency to subside. In contrast, some researchers have reported that rectangular-type cages often subside, despite the belief that this type of cage causes less subsidence. In a clinical study using the rectangular-type RABEA cage, subsidence greater than 2 mm in eight of 23 (35%) implants was observed, but this subsidence was not associated with clinical symptoms.12 Radiologically, the degree of subsidence in that study was unchanged at all affected levels at the 12-month follow-up compared with the follow-up at 3 months, indicating that the subsidence did not progress during that time, probably because of fusion. The authors suggested that over-distraction of the segment should be avoided to prevent cage subsidence. In another case series of 37 patients treated with the RABEA cage it was reported that cage subsidence of more than 2 mm occurred in 22% of patients, but that this was clinically not significant.13 Using the same type of cage as the one used in our study, Gercek et al. and found that five of the nine (56%) patients with fused levels had radiological signs of cage subsidence of 3 mm or more, and one patient had cage fracture.14 However, subsidence did not correlate with clinical symptoms. Based on our results, we can divide cage subsidence into two types. The first type, transient cage subsidence, is related to stabilization of the cage by the bony endplate subsequent to the interlocking of the cage spikes with the bone. This type of subsidence involved a distance of about 1–3 mm, occurred early in the postoperative period, and was not progressive. The second type, progressive cage subsidence, led to nonunion. In our opinion, risk factors for progressive cage subsidence are instability created by discectomy, postoperative cervical motion, cage design (wedgetype), oversized cage insertion with excessive distraction to obtain immediate stability, endplate preparation (excessive resection), and low bone mineral density. Improved cage design and size, materials, and initial stability are needed to reduce cage subsidence. In this series, an anatomicaltype cage was usually selected for simple disc herniation. However, a wedge-type cage was selected for disc herniation combined with spondylosis. To remove the posterior osteophyte completely, partial resection of the bony
1022
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
endplate was often required to maintain sufficient visualization. In terms of comparisons between the two types of cage, we noted that significant subsidence occurred only with the wedge-type cage in comparison with anatomicaltype cage. Partial resection of the bony endplate to insert the wedge-type cage might reduce initial stability, which relates to cage subsidence. We also found that cage size is important in preventing cage subsidence, because an oversized cage has a high rate of subsidence. New cage materials such as titanium, polylactic acid, allografts, polymethyl methacrylate, carbon, and polyetherether ketone (PEEK), will be important for reducing the risk of subsidence. The recently developed Solis cage (Stryker Spine, South Allendale, NJ, USA) is made of PEEK, a benzene ring polymer, and displays similar elastic behavior to cortical bone. Local autografted bone with the Solis cage has produced good clinical results and maintenance of segmental height, and local alignment was achieved during the follow-up period.15 Initial stability is also important for reducing cage subsidence. In an ex vivo study using human cadaveric spine segments, the biomechanical results for the WING cage, BAK/C cage, AcroMed cervical I/F cage, and bone cement were compared.16 The three cages acted as an alternative to bone cement in terms of providing the primary stabilizing effect in cervical interbody fusion.16 In that study, the initial stability of stand-alone interbody cages was compared with that of tricortical bone or the plate system. In flexion and extension, the plate had a significantly smaller range of motion than the cage and the autograft, and the cage had a significantly greater range of motion than the intact spine. This results of a another biomechanical study suggested that the cervical interbody cage should be supplemented with additional external or internal support to prevent excessive motion in flexion and extension.17 Several radiological parameters are used to assess the bony union after anterior cervical fusion, including abnormal motion in a dynamic lateral view, radiolucency between implant and endplate, the presence of bridging trabecular bone, and CT evaluation, although the best method remains a matter of debate.14 We used three radiological parameters to evaluate the status of the bony union. In a previous study, two radiographic parameters, the Cobb angle method and the spinous process method, were compared for assessment of pseudoarthrosis after ACDF.18 The operative segment was deemed fused if there was less than 2° of segmental movement on lateral flexion–extension views, if less than 50% of the anteroposterior distance of the interface between the endplates and implants was radiolucent, and if the interspinous process distance did not change more than 2 mm. The authors concluded that measuring the change in distance between spinous processes is more reproducible and accurate than the Cobb method for diagnosing pseudoarthrosis. In another previous study, 2° and 2 mm of motion were used as the upper limits to compensate for experimental error and variation.8
5. Conclusions In conclusion, two surgical procedures for cervical disc herniation were compared. Clinical results were almost the same. Cage method was less invasive than plate method including less donor site morbidities. High incidence of cage subsidence and loss of acquired alignment were recognized for cage group. Although stand-alone interbody cage had potential drawbacks radiologically, it was efficient alternate to conventional method. Surgeons must keep in mind the possible drawbacks of its subsidence. References 1. Smith GW, Robinson RA. The treatment of certain cervical-spine disorders by anterior removal of the intervertebral disc and interbody fusion. J Bone Joint Surg Am 1958;40:607–24. 2. Cloward RB. The anterior approach for removal of ruptured cervical disks. J Neurosurg 1958;15:602–17. 3. Banwart JC. Iliac crest bone graft harvest donor site morbidity. Spine 1995;20:1055–60. 4. Swain PD, Traynelis VC, Menezes AH. A comparative analysis of fusion rates and donor-site morbidity for autogenic rib and iliac crest bone grafts in posterior cervical fusions. J Neurosurg 1998;88:255–65. 5. Connolly PJ, Esses SI, Kostuik JP. Anterior cervical fusion: outcome analysis of patients fused with and without anterior cervical plates. J Spinal Disord 1996;9:202–6. 6. Lowery GL, McDonough RF. The significance of hardware failure in anterior cervical plate fixation. Patients with 2- to 7-year follow-up. Spine 1998;23:181–6. 7. Fujibayashi S, Shikata J, Kamiya N, et al. Missing anterior cervical plate and screws: a case report. Spine 2000;25:2258–61. 8. Hacker RJ, Cauthen JC, Gilbert TJ, et al. A prospective randomized multicenter clinical evaluation of an anterior cervical fusion cage. Spine 2000;25:2646–55. 9. Tureyen K. Disc height loss after anterior cervical microdiscectomy with titanium intervertebral cage fusion. Acta Neurochir 2003;145: 565–70. 10. Siddiqui AA, Jackowski A. Cage versus tricortical graft for cervical interbody fusion: a prospective randomized study. J Bone Joint Surg Br 2003;85-B:1019–25. 11. Matge G. Cervical cage fusion with 5 different implants: 250 cases. Acta Neurochir 2002;144:539–50. 12. Thome C, Krauss JK, Zevgaridis D. A prospective clinical comparison of rectangular titanium cages and iliac crest autografts in anterior cervical discectomy and fusion. Neurosurg Rev 2004;27:34–41. 13. Moreland DB, Asch HL, Clabeaux DE, et al. Anterior cervical discectomy and fusion with implantable titanium cage: initial impressions, patient outcomes and comparison to fusion with allograft. Spine J 2004;4:184–91. 14. Gercek E, Arlet V, Delisle J, et al. Subsidence of stand-alone cervical cages in anterior interbody fusion: warning. Eur Spine J 2003;12:513–6. 15. Shad A, Leach JCD, Teddy PJ, et al. Use of the Solis cage and local autologous bone graft for anterior cervical discectomy and fusion: early technical experience. J Neurosurg Spine 2005;2:116–22. 16. Wilke HJ, Kettler A, Claes L. Primary stabilizing effect of interbody fusion devices for the cervical spine: an in vitro comparison between three different cage types and bone cement. Eur Spine J 2000;9:410–6. 17. Shimamoto N, Cunningham BW, Dmitriev AE, et al. Biomechanical evaluation of stand-alone interbody fusion cages in the cervical spine. Spine 2001;26:E432–6. 18. Cannada LK, Scherping SC, Yoo JU, et al. Pseudoarthrosis of the cervical spine: a comparison of radiographic diagnostic measures. Spine 2003;28:46–51.
Journal of Clinical Neuroscience 15 (2008) 1017–1022 www.elsevier.com/locate/jocn
Clinical Study
Stand-alone interbody cage versus anterior cervical plate for treatment of cervical disc herniation: Sequential changes in cage subsidence Shunsuke Fujibayashi *, Masashi Neo, Takashi Nakamura Department of Orthopaedic Surgery, Graduate School of Medicine, Kyoto University, 54 Shogoin-Kawaharacho, Sakyo-ku, Kyoto 606-8507, Japan Received 10 April 2007; accepted 13 May 2007
Abstract Anterior cervical discectomy and fusion with an autogenous iliac bone graft is the gold standard treatment for cervical disc herniation. However, autologous bone grafts obtained from the anterior iliac crest are associated with significant donor-site morbidity and complications. To decrease bone graft-related problems, several types of interbody fusion cage have been developed and are used widely in clinical practice. We compared the clinical and radiological outcomes for two surgical procedures used to treat cervical disc herniation: the stand-alone interbody cage and autologous iliac bone grafting with an anterior plate. The clinical results did not differ between patients treated with the two procedures. The stand-alone cage was less invasive and had less donor-site morbidity. In patients treated with the bone graft and plate, the alignment of the fused segment was maintained in all but one patient, who exhibited nonunion. In contrast, in the cage-treated group, 44% of patients exhibited loss of lordotic alignment of more than 5° and cage subsidence of 3 mm or more. All cage subsidence occurred within 3 months of surgery. Although the stand-alone cage was a less invasive and more effective procedure to treat cervical disc herniation, surgeons should consider the possible drawbacks of the associated subsidence. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Cervical cage; Interbody fusion; Cervical disc herniation; Subsidence
1. Introduction Anterior cervical discectomy and fusion (ACDF) is the gold standard treatment for cervical disc herniation. Many technical modifications have been reported since its original description by Smith and Robinson1 and the later report by Cloward in 1958.2 Autologous iliac bone grafting is used widely either with or without a supplemental anterior plate. However, autologous bone grafts obtained from the anterior iliac crest are associated with significant donor-site morbidity and complications including severe acute pain, neurological injury, hematoma formation, infection and chronic pain.3,4 In patients treated without the plate, additional complications can include graft extrusion, collapse, and failure of fusion leading to kyphosis and *
Corresponding author. Tel.: +81757513362; fax: +81757518409. E-mail address: [email protected] (S. Fujibayashi).
0967-5868/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2007.05.011
pseudoarthrosis.5 The addition of an anterior plate system reduces the problem of graft extrusion and collapse but is itself associated with problems such as screw or plate dislodgement, dysphagia, and soft-tissue injury.6,7 To decrease the risk of such complications, several types of interbody fusion cage have been developed recently and are used widely in clinical practice. Cages are categorized roughly by shape into cylindrical and rectangular types. A sufficiently distracted intervertebral space after discectomy can be stabilized for multidirectional movement by tension forces of the residual annulus and ligaments. In clinical practice, distraction with restoration of disc height in a degenerative spine increases neuroforaminal volume and contributes to nerve root decompression. In this study, we used a stand-alone rectangular-type titanium cage to treat cervical disc herniation and compared the clinical and radiological results with those obtained using the conventional anterior plating method.
1018
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
2. Materials and methods
2.1. Surgical procedures
Eighteen consecutive patients with single-level cervical disc herniation that was refractory to conservative treatment were treated surgically. Nine patients were treated with the stand-alone cage procedure (cage group) by S.F. and another nine patients were treated with the anterior plating method (plate group) by M.N. In our institute, the treatment options for single-level cervical disc herniation include ACDF, posterior foraminotomy, and expansive laminoplasty. ACDF was selected for patients with myelopathy and/or radiculopathy with cord compression radiologically. Posterior foraminotomy was selected for patients with radiculopathy without cord compression; and expansive laminoplasty was selected for patients with disc herniation combined with symptomatic developmental spinal canal stenosis. One of the 18 patients with failed posterior foraminotomy was successfully treated using ACDF. The demographic and clinical data were similar in the two groups (Table 1). The mean age at the time of surgery was 53.3 years (range, 31–75 years) in the cage group and 49.0 years (28–64 years) in the plate group. The mean follow-up period was 23.1 months (12–30 months) in the cage group and 25.2 months (13–36 months) in the plate group. Preoperatively, 11 patients had myelopathy and seven had radiculopathy. Table 2 summarizes the levels involved. The most common cervical level involved was C5–6. The operative time and intraoperative blood loss were compared to assess the surgical invasiveness for the two methods. The Japanese Orthopaedic Association (JOA) score and recovery rate (Hirabayashi’s method) were compared before surgery, 3 months after surgery, and at the latest follow-up examination for each group. Donor-site pain was assessed at 3 months and the latest follow-up examination.
Surgical procedures were performed using a standard left anterolateral approach. After insertion of a cervical spine distracter, complete discectomy and neural decompression were performed using a surgical microscope. The cartilaginous endplate was removed completely to expose the cortical endplate. The bony endplate was preserved as much as possible to prevent cage subsidence. An appropriate-sized cage (Syncage-C; Synthesis, Paoli, PA, USA) was filled with autologous cancellous bone harvested as a cylinder from the left anterior iliac crest through a mini-incision using a special device. The cage size was determined by both preoperative templating and intraoperative evaluation using a trial cage to confirm initial stability. The cage was inserted into the disc space by using an impactor and cage stability was confirmed after the distracter was removed. Patients remained in a soft collar for 4 weeks postoperatively. Anterior plating was performed according to the Smith–Robinson technique, in which tricortical autologous iliac bone is harvested from the iliac crest and grafted between the vertebral bodies under manual traction. Supplemental anterior plate fixation was applied using an Atlantis plate system (Medtronic Sofamor Danek, Minneapolis, MN, USA).
Table 1 Patient demographic data
Age (years) Male/female Follow-up period (months) Myelopathy/radiculopathy
Cage group
Plate group
Significance
53.3 (31–75) 3/6 23.1 (12–30) 5/4
49.0 (28–64) 6/3 25.2 (13–36) 6/3
NS NS
Statistical analysis was performed using Student’s t-test and significance was set at p < 0:05. NS, not significant.
Table 2 Levels involved Level
No. patients
C3/4 C4/5 C5/6 C6/7
2 4 7 5
2.2. Radiological assessments To assess bony union, three different radiological parameters on lateral dynamic radiographs were determined at several times: before surgery, immediately after surgery, every 3 months after surgery, and at the latest follow-up examination. The fused segment angle (FSA) was defined as the angle formed between the lines drawn parallel to the cranial endplate of the cranial vertebrae and the caudal endplate of the caudal vertebrae. More than 2° motion at flexion–extension was considered to indicate nonunion. The interspinous process distance at flexion– extension was measured between the tips of both spinous processes. More than 2 mm motion at flexion–extension was considered to indicate nonunion. In addition, radiolucency >50% over the anteroposterior distance of the interface between the endplates and implants was defined as nonunion. Multidirectional CT assessment of the coronal and sagittal reconstruction views was also performed 6 and 12 months after surgery. Union was considered to have occurred when bony trabecular orientation was visible (Fig. 1). Radiologically successful fusion was considered to have occurred when all of the three radiological parameters mentioned above and CT assessments indicated fusion. Cage subsidence was calculated from the change of fused segment height (FSH), which was ascertained using the lengths of lines drawn between the center of the cranial endplate of the cranial vertebrae and the center of the caudal end of the caudal vertebrae. A change of 3 mm or more was defined as significant cage subsidence.
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
Fig. 1. Sagittal (left) and coronal (right) reconstructed CT images of the cervical vertebrae of a patient who underwent interbody cage placement for cervical disc herniation. A continuous bony trabecular pattern can be seen through the cage.
Statistical analysis was performed using Student’s t-test; p < 0:05 was accepted as significant. 3. Results The clinical results are summarized in Table 3. Although not significant, the level of surgical invasiveness was lower in the cage group than in the plate group. The surgical time was 124.4 min in the cage group and 149.7 min in the plate group. The intraoperative blood loss was 13.6 mL in the cage group and 19.2 mL in the plate group (not significant, NS). The JOA recovery rate was 96.2% in the cage group and 79.3% in the plate group (NS). Three months after surgery, 22.2% of patients in the plate group complained of donor-site pain compared with none in the cage group. No patient in either group complained of donor-site pain at the final follow-up. We observed no major complications related to surgical interventions in either group throughout the follow-up periods. Radiologically successful fusion was achieved by 88.8% of patients in both groups. The mean time until bony union was 5.6 months (3–18 months) in the cage group and 7.7 months (5–12 months) in the plate group (NS).
Table 3 Clinical and radiological results
Operative time (min) Blood loss (mL) JOA score recovery rate (%) Donor site pain (%) At 3 months At latest follow-up Bony union (%) Time until bony union (months)
Cage group
Plate group
Significance
124.4 (82–181) 13.6 (0–61) 96.2 (73.7–100)
149.7 (130–200) 19.2 (0–63) 79.3 (59.3–100)
NS NS NS
1019
In the cage group, the mean FSA was –1.78° before surgery, 3.89° immediately after surgery, 1.67° 3 months after surgery, and –0.78° at the final follow-up examination. The FSA values differed significantly between before surgery and immediately after surgery (p ¼ 0:0041), between immediately after surgery and 3 months after surgery (p ¼ 0:0304), and between 3 months after surgery and the final follow-up examination (p ¼ 0:0237). The difference in FSA between before surgery and the final follow-up examination was not significant. In the cage group, the mean FSH was 33.11 mm before surgery, 34.67 mm immediately after surgery, 33.33 mm 3 months after surgery, and 32.78 mm at the final followup examination. The FSH values differed significantly between before and immediately after surgery ðp ¼ 0:0278Þ, and between immediately after and 3 months after surgery ðp ¼ 0:0114Þ. The differences in FSA between before surgery and the final follow-up examination, and between 3 months after surgery and the final follow-up examination, were not significant. In contrast, in the plate group, the mean FSA was 3.22° before surgery, 0.22° immediately after surgery, 0.22° 3 months after surgery, and 0.33° at the final follow-up examination. The FSA differed significantly between before and immediately after surgery ðp ¼ 0:0122Þ and between before surgery and the final follow-up examination ðp ¼ 0:0160Þ. In the plate group, the mean FSH was 35.22 mm before surgery, 36.22 mm immediately after surgery, 35.67 mm 3 months after surgery, and 35.22 mm at the final follow-up examination. Only the difference between the values before and immediately after surgery was significant ðp ¼ 0:0085Þ. The acquired FSH gradually decreased to presurgery values, although this was not significant. These results indicate that, although FSH was not maintained in the plate group, the FSA was corrected and maintained through the follow-up period. Overall, the alignment of the fused segment was maintained well in all but one case of nonunion in the plate group. In contrast, a loss of lordotic alignment of >5° occurred in 44% of patients (Fig. 2a,b), and cage subsidence of P3 mm occurred in 44% of cases (Fig. 3a,b). All cage subsidence occurred within 3 months of surgery, but no progressive subsidence was observed after this time. Interestingly, all cages showed about 1–2 mm of minimum subsidence in the early postoperative period. No anterior or posterior migration was observed. Fig. 4 presents the results of a typical successful union and Fig. 5 a typical nonunion with progressive cage subsidence. 4. Discussion
0 0 88.8 5.6 (3–18)
22.2 0 88.8 7.7 (5–12)
NS
Statistical analysis was performed using Student’s t-test and significance was defined at p < 0:05. JOA, Japanese Orthopaedic Association; NS, not significant.
The possible advantages of the stand-alone interbody cage are that it is less invasive; requires a simple surgical procedure, short operating time, and smaller amount of harvested bone; causes less blood loss; is associated with a shorter period of postoperative external immobilization; and can correct cervical kyphosis and restore disc height.
1020
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
Fig. 2. Sequential change in the fused segment angle (FSA) in the cage (a) and plate (b) groups. The loss of alignment was >5° in 44% of patients in the cage group (a), but alignment was maintained well in the plate group (b) throughout the follow-up period. Pre-op, preoperative; Post-op, postoperative; 3 months, 3-month follow-up; Latest F/U, latest follow-up examination.
Fig. 3. Sequential changes in fused segment height (FSH) in the cage (a) and plate (b) groups. Forty-four percent of patients in the cage group (a) showed a decrease in FSH of P3 mm, which occurred mainly within the 3 months following the operation. In contrast, FSH was maintained well in the plate group (b) during the follow-up period. Pre-op, preoperative; Post-op, postoperative; 3 months, 3-month follow-up; Latest F/U, latest follow-up examination.
Fig. 4. Sequential lateral X-rays of the cervical vertebrae of a patient with successful fusion immediately after, 3 months after, and 1 year after surgery for interbody cage placement. Cage subsidence of 1 mm with interlocking with the cage and vertebral endplate were recognized at 3 months and no progression of cage subsidence was evident at the 1 year follow-up. Post-op, postoperative.
Although the present study had some limitations, including a small sample size and no patient randomization, our results showing significant cage subsidence and loss of cervical alignment show that there are potential disadvantages as well as possible advantages. In the cage procedure, correction of operative alignment and restoration of disc height were achieved more effec-
tively and easily than with the conventional approach. However, the acquired alignment and height tended to return gradually toward presurgical values. In contrast, plating maintained the acquired lordotic alignment to achieve successful bony union, although disc height tended to return gradually to the presurgical value. These results could be attributed to surgical procedures, such as the use of a
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
1021
Fig. 5. Sequential lateral X-rays of the cervical vertebrae of a patient with nonunion, immediately after, 3 months after, and 1 year after surgery interbody cage placement. An oversized wedge-shaped cage was inserted with excessive distraction. Cage subsidence of 4 mm was recognized at 3 months, and further cage subsidence and a radiolucent line between the cage and endplate was observed at the 1 year follow-up. Post-op, postoperative.
special distracter and impaction device in the cage group, which increase the disc height during surgery. We found it difficult to maintain the acquired height in the longitudinal direction in either procedure. Although effective disc and foraminal height was restored by surgery, this restoration was lost gradually until the posterior facet joints were stabilized. The stand-alone cage has been used widely in clinical practice, and successful clinical results have been reported in more than one study, although the reliability of this technique remains controversial. In a multicenter study undertaken to obtain US Federal Drug Administration approval, in which the cylindrical cage (BAK/C; Sulzer Spine-Tech Minneapolis, MN, USA) was compared with noninstrumented bone-only fusion, similar success rates were achieved for the two techniques.8 In the multicenter study, the complication rate associated with ACDF was 20%, which was lower than the overall complication rate of 12% for the cage. In a case-series involving 47 patients, 98% of patients treated with the BAK/C cage had achieved solid fusion at an average of 6 months after the operation, but the use of a cervical intervertebral cage in anterior cervical microdiscectomy did not prevent a reduction in the height of the cervical disc space after surgery.9 In another study in which the cage was compared with a tricortical graft it was shown that tricortical graft fusion is cheaper than cage fusion and is more effective in reducing the pain score.10 The shape of the implant and preparation of the endplates both influence the tendency for cage subsidence of an intervertebral implant. A large contact surface together with intact endplates decreases the tendency for subsidence, whereas a small surface area and destructive preparation of the end plates increases the risk of subsidence. In a large comparative clinical study of the BAK/C cage, WING cage (Medinorm, Quierschied, Germany), RABEA cage (Signus, Alzenau, Germany), NOVUS cage (Sofamor Danek, Memphis TN, USA), and CBK cage (Scient’X, Paris, France), asymptomatic anterior cage migration and posterior intracanal migration were associated with cylindrical devices.11 Significant subsidence was observed with cylindrical devices, although minor subsidence of 1–2 mm was observed frequently with the other devices. The authors
concluded that rectangular cages allow for better preservation of sagittal balance or correction because of a lesser tendency to subside. In contrast, some researchers have reported that rectangular-type cages often subside, despite the belief that this type of cage causes less subsidence. In a clinical study using the rectangular-type RABEA cage, subsidence greater than 2 mm in eight of 23 (35%) implants was observed, but this subsidence was not associated with clinical symptoms.12 Radiologically, the degree of subsidence in that study was unchanged at all affected levels at the 12-month follow-up compared with the follow-up at 3 months, indicating that the subsidence did not progress during that time, probably because of fusion. The authors suggested that over-distraction of the segment should be avoided to prevent cage subsidence. In another case series of 37 patients treated with the RABEA cage it was reported that cage subsidence of more than 2 mm occurred in 22% of patients, but that this was clinically not significant.13 Using the same type of cage as the one used in our study, Gercek et al. and found that five of the nine (56%) patients with fused levels had radiological signs of cage subsidence of 3 mm or more, and one patient had cage fracture.14 However, subsidence did not correlate with clinical symptoms. Based on our results, we can divide cage subsidence into two types. The first type, transient cage subsidence, is related to stabilization of the cage by the bony endplate subsequent to the interlocking of the cage spikes with the bone. This type of subsidence involved a distance of about 1–3 mm, occurred early in the postoperative period, and was not progressive. The second type, progressive cage subsidence, led to nonunion. In our opinion, risk factors for progressive cage subsidence are instability created by discectomy, postoperative cervical motion, cage design (wedgetype), oversized cage insertion with excessive distraction to obtain immediate stability, endplate preparation (excessive resection), and low bone mineral density. Improved cage design and size, materials, and initial stability are needed to reduce cage subsidence. In this series, an anatomicaltype cage was usually selected for simple disc herniation. However, a wedge-type cage was selected for disc herniation combined with spondylosis. To remove the posterior osteophyte completely, partial resection of the bony
1022
S. Fujibayashi et al. / Journal of Clinical Neuroscience 15 (2008) 1017–1022
endplate was often required to maintain sufficient visualization. In terms of comparisons between the two types of cage, we noted that significant subsidence occurred only with the wedge-type cage in comparison with anatomicaltype cage. Partial resection of the bony endplate to insert the wedge-type cage might reduce initial stability, which relates to cage subsidence. We also found that cage size is important in preventing cage subsidence, because an oversized cage has a high rate of subsidence. New cage materials such as titanium, polylactic acid, allografts, polymethyl methacrylate, carbon, and polyetherether ketone (PEEK), will be important for reducing the risk of subsidence. The recently developed Solis cage (Stryker Spine, South Allendale, NJ, USA) is made of PEEK, a benzene ring polymer, and displays similar elastic behavior to cortical bone. Local autografted bone with the Solis cage has produced good clinical results and maintenance of segmental height, and local alignment was achieved during the follow-up period.15 Initial stability is also important for reducing cage subsidence. In an ex vivo study using human cadaveric spine segments, the biomechanical results for the WING cage, BAK/C cage, AcroMed cervical I/F cage, and bone cement were compared.16 The three cages acted as an alternative to bone cement in terms of providing the primary stabilizing effect in cervical interbody fusion.16 In that study, the initial stability of stand-alone interbody cages was compared with that of tricortical bone or the plate system. In flexion and extension, the plate had a significantly smaller range of motion than the cage and the autograft, and the cage had a significantly greater range of motion than the intact spine. This results of a another biomechanical study suggested that the cervical interbody cage should be supplemented with additional external or internal support to prevent excessive motion in flexion and extension.17 Several radiological parameters are used to assess the bony union after anterior cervical fusion, including abnormal motion in a dynamic lateral view, radiolucency between implant and endplate, the presence of bridging trabecular bone, and CT evaluation, although the best method remains a matter of debate.14 We used three radiological parameters to evaluate the status of the bony union. In a previous study, two radiographic parameters, the Cobb angle method and the spinous process method, were compared for assessment of pseudoarthrosis after ACDF.18 The operative segment was deemed fused if there was less than 2° of segmental movement on lateral flexion–extension views, if less than 50% of the anteroposterior distance of the interface between the endplates and implants was radiolucent, and if the interspinous process distance did not change more than 2 mm. The authors concluded that measuring the change in distance between spinous processes is more reproducible and accurate than the Cobb method for diagnosing pseudoarthrosis. In another previous study, 2° and 2 mm of motion were used as the upper limits to compensate for experimental error and variation.8
5. Conclusions In conclusion, two surgical procedures for cervical disc herniation were compared. Clinical results were almost the same. Cage method was less invasive than plate method including less donor site morbidities. High incidence of cage subsidence and loss of acquired alignment were recognized for cage group. Although stand-alone interbody cage had potential drawbacks radiologically, it was efficient alternate to conventional method. Surgeons must keep in mind the possible drawbacks of its subsidence. References 1. Smith GW, Robinson RA. The treatment of certain cervical-spine disorders by anterior removal of the intervertebral disc and interbody fusion. J Bone Joint Surg Am 1958;40:607–24. 2. Cloward RB. The anterior approach for removal of ruptured cervical disks. J Neurosurg 1958;15:602–17. 3. Banwart JC. Iliac crest bone graft harvest donor site morbidity. Spine 1995;20:1055–60. 4. Swain PD, Traynelis VC, Menezes AH. A comparative analysis of fusion rates and donor-site morbidity for autogenic rib and iliac crest bone grafts in posterior cervical fusions. J Neurosurg 1998;88:255–65. 5. Connolly PJ, Esses SI, Kostuik JP. Anterior cervical fusion: outcome analysis of patients fused with and without anterior cervical plates. J Spinal Disord 1996;9:202–6. 6. Lowery GL, McDonough RF. The significance of hardware failure in anterior cervical plate fixation. Patients with 2- to 7-year follow-up. Spine 1998;23:181–6. 7. Fujibayashi S, Shikata J, Kamiya N, et al. Missing anterior cervical plate and screws: a case report. Spine 2000;25:2258–61. 8. Hacker RJ, Cauthen JC, Gilbert TJ, et al. A prospective randomized multicenter clinical evaluation of an anterior cervical fusion cage. Spine 2000;25:2646–55. 9. Tureyen K. Disc height loss after anterior cervical microdiscectomy with titanium intervertebral cage fusion. Acta Neurochir 2003;145: 565–70. 10. Siddiqui AA, Jackowski A. Cage versus tricortical graft for cervical interbody fusion: a prospective randomized study. J Bone Joint Surg Br 2003;85-B:1019–25. 11. Matge G. Cervical cage fusion with 5 different implants: 250 cases. Acta Neurochir 2002;144:539–50. 12. Thome C, Krauss JK, Zevgaridis D. A prospective clinical comparison of rectangular titanium cages and iliac crest autografts in anterior cervical discectomy and fusion. Neurosurg Rev 2004;27:34–41. 13. Moreland DB, Asch HL, Clabeaux DE, et al. Anterior cervical discectomy and fusion with implantable titanium cage: initial impressions, patient outcomes and comparison to fusion with allograft. Spine J 2004;4:184–91. 14. Gercek E, Arlet V, Delisle J, et al. Subsidence of stand-alone cervical cages in anterior interbody fusion: warning. Eur Spine J 2003;12:513–6. 15. Shad A, Leach JCD, Teddy PJ, et al. Use of the Solis cage and local autologous bone graft for anterior cervical discectomy and fusion: early technical experience. J Neurosurg Spine 2005;2:116–22. 16. Wilke HJ, Kettler A, Claes L. Primary stabilizing effect of interbody fusion devices for the cervical spine: an in vitro comparison between three different cage types and bone cement. Eur Spine J 2000;9:410–6. 17. Shimamoto N, Cunningham BW, Dmitriev AE, et al. Biomechanical evaluation of stand-alone interbody fusion cages in the cervical spine. Spine 2001;26:E432–6. 18. Cannada LK, Scherping SC, Yoo JU, et al. Pseudoarthrosis of the cervical spine: a comparison of radiographic diagnostic measures. Spine 2003;28:46–51.