Spinal fusion has been used for decades to manage a variety of spinal disorders. The earliest fusions typically involved a large amount of muscle dissection, copious amounts of autogenous bone graft, and periods of enforced bed rest. Spinal instrumentation was introduced to increase the rate of successful fusion, decrease the period of patient recovery, and allow surgeons to alter the position of the spine. To minimize failure rates of posterior instrumentation, concepts of anterior column support were developed. Originally this necessitated bone being placed in the disc space either from an anterior or posterior approach. This had the advantage of a better blood supply, weight-bearing fusion, and better ability to maintain lordosis. However, when bone graft was used as a stand-alone device, it did not prove to be strong enough. Bone-only anterior lumbar interbody fusion (ALIF) and posterior lumbar interbody fusion (PLIF) had a significant incidence of collapse and pseudarthrosis. Intervertebral cages or spacers were developed to prevent the collapse and pseudarthrosis seen with bone-only fusions. Vertically placed titanium mesh cylinders and rectangular carbon fiber cages were among the earliest spacers used for this reason.
Bagby and Kuslich ushered in the era of stand-alone threaded intervertebral cages. Much of this work was based on Bagby’s initial use of a cylindrical metal basket to facilitate equine cervical fusions. Theoretically, threaded intervertebral cages would stabilize a segment through distraction and tensioning of the anular and ligamentous structures. Then by partially reaming the endplates, cancellous bone would be exposed for arthrodesis. The threaded interface was proposed to stabilize the segment adequately and preclude supplemental fixation. Since the introduction of these cages, the field of interbody fusion technology has literally exploded and generated a great deal of interest and controversy.
Initial biomechanical studies suggested that threaded lumbar intervertebral cages are stable enough to be used as stand-alone devices. In an early study, Brodke et al showed that two threaded BAK cages placed using a posterior approach resulted in greater stability of the calf spine motion segment than a PLIF bone graft alone, and resulted in equivalent stability to a PLIF bone graft and pedicle screw construct. 1 They also noted that the bone graft stand-alone PLIF was less stable than the intact specimen. In human spines, insertion of threaded PLIF cages results in reduced torsional stiffness as a result of the facet joint resection required for the appropriately sized cage insertion. 2 Kettler et al found that posteriorly placed intervertebral cages stabilized the spine only in flexion and lateral bending. No increased stability was found in extension and rotation. In addition, reduced stability under cyclical loading (40,000 cycles) conditions was found. 3
Oxland et al4 reported that anteriorly placed threaded cylindrical cages did enhance motion segment stability in all loading motions except extension. They found that supplementary translaminar facet screw fixation provided additional stabilization only in extension. Similarly, Rathonyi et al5 found that translaminar facet screw fixation greatly improved the stability of anteriorly placed threaded cylindrical cages in extension and rotation. The direction of cage insertion has also been studied biomechanically. It appears that a single oblique cage or a single laterally placed cage offers similar immediate biomechanical stability as two anteriorly placed devices. 2 However, single cages do not offer the same surface area for arthrodesis as two cages.
Initial threaded cage stability is thought to be dependent on achieving appropriate disc space distraction that will result in anular tensioning. Although the concept of anular tensioning has been emphasized, it has not been well studied, and clinical study has shown that complete removal of the anterior anulus does not adversely affect fusion rates. 6 Others have suggested that threaded cage stability is at least in part dependent on compressive preload generated by muscle forces across the disc space. In a biomechanical study, Phillips et al7 demonstrated that during low preload conditions, corresponding to those preloads experienced during supine and recumbent postures, anteriorly placed BAK cages were less effective at stabilizing the motion segment in extension. Higher compressive loads corresponding to those generated during sitting or standing significantly stabilized the motion segment in all motion planes. Supplemental posterior stabilization with translaminar facet screws did provide a significant stabilizing advantage during low preload conditions. Another important factor that affects stability of threaded cages is vertebral body bone density. Lund et al8 showed a significant relationship between bone density of the adjacent vertebral bodies and stability of the interbody cage construct. Second-generation lumbar cages have included the lumbar tapered cage (LT Cage, Medtronic, Minneapolis, MN), which is purported to be endplate sparing and able to obtain lordosis by threading a wedge-shaped device into the disc space.
Shimamoto et al9 demonstrated that in flexion and extension, a threaded cervical cage (RABEA) did not offer stiffness equivalent to the intact spine. An anterior locking plate demonstrated significantly lower range of motion than did the cage or a tricortical bone graft in flexion, extension, and axial rotation. The authors recommended additional cervical plating to provide adequate cage stability. Kandziora et al10 reported that flexion stiffness increased after threaded cervical cage insertion compared with that of the intact motion segment, whereas rotation stiffness decreased after cage implantation. They also reported that cylindrical devices were able to control extension and bending more effectively than threaded cages. In a biomechanical study using human cervical spine specimens, Wilke et al11 reported significant subsidence of the BAK/C threaded cage after applying loading cycles that simulated the patient’s neck movements during the first postoperative days.
The biology of fusion with the use of interbody cages in humans has largely been extrapolated from animal studies. In a study of vertebral specimens from horses that had anterior interbody arthrodesis with a stainless steel Bagby basket, Cunningham et al12 reported successful fusion with mature trabecular bone spanning the disc space at 4 years. There was a significant decrease in bone density at the fusion site compared with that of the adjacent vertebral bodies that was thought to result from stress shielding. The results of needle biopsies obtained from tissue within radiographically successful interbody fusion cages in eight patients were reported recently. Biopsies showed fragments of necrotic bone associated with viable bone and restoration of hematopoietic bone marrow. Small debris particles were seen in carbon fiber and titanium cages, but there was no visible bone resorption or inflammation. 13 Fusion biology seems to also be impacted by the cage design. In a sheep spine model, Kandziora et al14 showed significantly higher bone mineral and bone callus content with threaded cylindrical intervertebral cages compared with titanium box design cages and autologous tricortical iliac crest. Polychrome sequential labeling showed more advanced bone matrix formation in the cylindrical cage group. These findings emphasize the importance of adequate study of any new cage device because the bone healing response cannot necessarily be assumed.
The initial Food and Drug Administration (FDA) trials for the threaded cylindrical cages showed promising radiographic and clinical results. Ray 15 reported a 96% fusion rate at 2-year follow-up evaluation, with 65% having good or excellent results, 21% fair, and 14% poor. Kuslich et al16 reported their 4-year follow-up results of a select group of patients from their FDA Investigational Device Exemptions (IDE) trial. They stated a 95.1% fusion rate and found that 63% of patients were gainfully employed. They noted an overall complication rate of 13.8%, with 8.7% undergoing a second operation. The IDE data leading to LT Cage approval demonstrated 93% successful fusions at 2 years, with 72% overall patient satisfaction. 17
Others, however, have been unable to duplicate these high success rates. Elias et al18 reported on a series of 67 patients who underwent posterior interbody threaded fusion cages. They reported a high complication rate with 15% dural tears, 5% unilateral cage placement, 15% postoperative radiculopathy, and 30% pseudarthrosis. Twenty-one percent of patients underwent revision surgery. O’Dowd et al, 19 using laparoscopic and open anterior approaches, showed an unacceptably high rate of complications and pseudarthrosis using stand-alone cylindrical cages in 48 consecutive patients. A high incidence of approach-related complications was reported, and of the 22 laparoscopic cases, complications included four major vessel injuries, one wound infection, one colon injury, and eight patients with L5 root pain. Thirty-one percent of patients underwent revision surgery. The authors’ selection criteria have been questioned because a number of patients in this study had had previous surgery, including laminectomy, and might not be considered ideal stand-alone anterior interbody fusion candidates. Heim and Abitbol 20 have detailed strategies for the salvage of failed intervertebral fixation devices. Many of these failures are the result of technical difficulties and poor patient selection.
One of the concerns in interpreting the reported results for interbody fusion surgery is the difficulty in determining fusion success radiographically. This is particularly problematic when metal devices are used. Most published reports use the “fusion” criteria required by the FDA, including motion of less than 5° on flexion–extension radiographs and an absence of lucencies around the cages or cage migration. Although meeting these criteria suggests the cage is stable, it does not necessarily indicate true fusion of the motion segment. Using these criteria, fusion rates are likely overestimated. McAfee et al21 have reported the unreliability of assessment of fusion based on flexion and extension radiographs. Computed tomography (CT) scans have been suggested as an alternative technique for assessing interbody fusion; however, studies assessing the accuracy of this technique are lacking.
What has become clear during the 10-year experience using threaded intervertebral cages is that patient selection and technical expertise are equally important. Ideal selection criteria include anatomic and psychological factors. Ideal anatomy would be a single level disc space with collapse, endplate osteophytes, and sclerosis. In these patients, a surgeon can expect an acceptable degree of clinical success from stand-alone cage procedures. However, when patients with multilevel cases are operated on, when disks have a normal (tall) appearance on radiographs, or when patients have osteoporotic bone, threaded intervertebral cage devices have been shown to have a higher incidence of complication and failure. In patients with spondylolisthesis or previous laminectomy where abnormal motion or potential for instability exists, posterior instrumentation to supplement the interbody construct should be considered. Clearly, patients with chronic pain behaviors or substance abuse disorders are not surgical candidates.
A variety of approaches to the spine have been used for the placement of intervertebral cages. Anteriorly, open approaches, usually retroperitoneal, and laparoscopic approaches, usually transperitoneal, have been used. Although the laparoscopic approach to the anterior spine has been popularized, the long learning curve and the potential complications from this approach have been documented. 22–24 Clearly, the laparoscopic approach to L5–S1 is feasible, although in male patients a higher incidence of retrograde ejaculation will result. At L4–L5, it is clear that the laparoscopic approach becomes much more difficult. The ability to ligate and control the iliolumbar vein and to retract the great vessels across the spine at L4–L5 is limited using laparoscopic technology. For general use, the mini-open retroperitoneal approach at L4–L5 has been shown to have a lower complication rate and a similar postoperative recovery time as the laparoscopic approach. 25
Posteriorly, the use of stand-alone threaded intervertebral cages has decreased. This has been the result of better understanding the destabilizing effect of facet resection required for cage placement and also because of the degree of neural retraction required to place appropriately sized threaded cages. Supplemental pedicle screw fixation greatly enhances the stability of intervertebral devices. In addition, lordosis can only be restored when cages and screws are used in conjunction. 26 If pedicle screws are used, it may not be necessary to use the higher technology threaded cages. Harms pioneered the use of upright titanium mesh spacers in the disc space. He used a unilateral transforaminal interbody fusion (TLIF) approach for the placement of these cages. This approach has been shown to be successful by numerous authors. In a series of 35 patients, Whitecloud et al27 showed a 97% fusion rate using the TLIF approach, pedicle screws, and titanium spacers. Similarly, in a series of 22 patients with degenerative spondylolisthesis, Rosenberg et al28 demonstrated a high degree of clinical and radiographic success using this approach. The PLIF approach, which is a true posterior interbody fusion with bilateral cages, appears to have a slightly higher complication rate. Agazzi et al29 examined the use of PLIF in 71 consecutive patients. Although they stated a 90% fusion rate, their overall satisfaction rate of 67% and their 39% good or excellent results are less than satisfactory. Others, however, remain steadfast proponents of the PLIF approach. In a series of 178 patients, Brantigan et al30 reported a fusion success rate of 98.9% and a clinical success rate of 86% using carbon fiber PLIF cages with pedicle screw fixation.
It certainly appears that interbody spacers placed from a posterior or lateral approach have their place in spinal surgery. Particularly for patients with spondylolisthesis where pedicle screw fixation is being used, intervertebral spacers tend to yield a better fusion rate, better spine position, and good clinical results. When treating patients with back pain from degenerative disc disease, however, the clinical results do not increase to the level of the radiographic fusion successes, and it is unclear if the addition of intervertebral spacers to the pedicle screw construct improves clinical results.
Cervical Intervertebral Cages
Similar threaded intervertebral cage technology has been used in the cervical spine. Hacker et al31 have presented their results of a prospective randomized multicenter clinical trial using a threaded cage (BAK/C) for the management of “cervical diskogenic radiculopathy.” They found equivalent clinical results with the BAK/C compared with autogenous tricortical iliac crest bone graft. They reported a successful arthrodesis rate of 98% for one-level BAK/C procedures and a complication rate that was lower than that observed with standard uninstrumented anterior cervical discectomy and fusion (ACDF) techniques. Profeta et al32 have used a threaded tapered cervical cage device (Novus-CT; Affinity; Medtronic; Memphis, TN) and similarly found high fusion and clinical success rates. Matge 33 reported on 250 consecutive patients treated with threaded titanium cervical cages (149 patients), impacted polyetheretherketone (PEEK) cages (59), or impacted titanium cages (42 cases) with local bone graft or bone substitute. Excellent outcome for neck pain and radiculopathy was reported, but less favorable outcome for myelopathy was observed. Cage complications included migration and subsidence. Although the clinical applications may be limited, cervical cages may offer an alternative to standard graft and plate techniques.
Recent advances in cages include changes in cage design, cage materials, and osteoinductive materials placed within cages. Certainly the combination of recombinant human bone morphogenetic protein-2 (rhBMP-2) with the titanium tapered threaded cage (LT Cage plus InFuse, Medtronic, Minneapolis, MN) has met with a high degree of clinical and radiographic success in its recent FDA trial. 17 We will leave the discussion of rhBMP-2 to a separate report. Current research also includes the use of cages for patients with idiopathic scoliosis, the use of resorbable cages, and the use of radiolucent cages. 34 Clinical studies are currently underway examining the combination of resorbable cages and bone growth factors for clinical use.
The inability to radiographically assess fusion with intervertebral cages is an inherent limitation of many currently used devices. This has given some impetus to the development of biologic spacers that will provide immediate spinal stability but will ultimately be replaced by or incorporated into the healed fusion (biodegradable). This should allow for better assessment of fusion using conventional radiographic techniques. Machined allograft bone in the form of wedges or threaded dowels has been used as a biologic spacer. Conflicting clinical results have been reported with these devices, and it appears that determining fusion radiographically is not any easier than with metal devices. In addition, allograft bone may take many years to be incorporated and may never be completely replaced by host bone. Early testing of biodegradable polylactic acid cages has been reported. 35,36 These cages have reduced stiffness compared with metal devices that proponents believe may reduce device-related osteopenia. van Dijk et al36 compared a polylactic acid resorbable cage with titanium cages in a goat model. They found that the reduced stiffness of the resorbable cages enhanced the interbody fusion compared with titanium cages. Although early research into alternate biomaterial for cage technology is promising, the optimal cage stiffness and the desired period over which the cage will biodegrade remain unknown.
Clearly the introduction of intervertebral cages during the past two decades has generated a great deal of interest and controversy. Although the use of disc reparative techniques and disc arthroplasty may limit the future use of cages, they are currently commonly used in spinal surgery. For the management of single-level degenerative discs, anteriorly placed intervertebral threaded cages have been shown to have a high degree of clinical and radiographic success. Similarly, the use of posteriorly or transforaminally placed cages within the discs, supplemented by pedicle screws, has reproducibly led to a high degree of clinical success for spondylolisthesis. It is with other indications, though, that cage use is somewhat more questionable. For the management of pure discogenic pain without collapse, multilevel disc disease, or disc degeneration in elderly patients, the use of cages is certainly less certain. It is for these difficult patient groups that future research will help us delineate the truly successful procedures.
- Interbody spinal fusion eases work well in the treatment of disc degeneration.
- Patient selection is critical to ensure good results.
1. Brodke DS, Dick JC, Kunz DN, et al. Posterior lumbar interbody
fusion. A biomechanical comparison, including a new threaded cage
. Spine 1997; 22: 26–31.
2. Tencer AF, Hampton D, Eddy S. Biomechanical properties of threaded inserts for lumbar interbody
spinal fusion. Spine 1995; 20: 2408–14.
3. Kettler A, Wilke HJ, Dietl R, et al. Stabilizing effect of posterior lumbar interbody
fusion cages before and after cyclic loading. J Neurosurg
4. Oxland TR, Hoffer Z, Nydegger T, et al. A comparative biomechanical investigation of anterior lumbar interbody
cages: Central and bilateral approaches. J Bone Joint Surg Am 2000; 82A: 383–93.
5. Rathonyi GC, Oxland TR, Gerich U, et al. The role of supplemental translaminar screws in anterior lumbar interbody
fixation: A biomechanical study. Eur Spine J 1998; 7: 400–7.
6. McAfee PC, Lee GA, Fedder IL, et al. Anterior BAK instrumentation and fusion: Complete versus partial diskectomy. Clin Orthop 2002; 394: 55–63.
7. Phillips FM, Cunningham B, Patwardhan AG, et al. Effects of supplemental translaminar facet screw fixation on the stability of stand alone ALIF cages under physiologic compressive preload. In press.
8. Lund T, Oxland TR, Jost B, et al. Interbody cage
stabilization in the lumbar spine. Biomechanical evaluation of cage
design, posterior instrumentation and bone density. J Bone Joint Surg Br 1998; 80: 351–9.
9. 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.
10. Kandziora F, Pflugmacher R, Schafer J, et al. Biomechanical comparison of cervical interbody
fusion cages. Spine 2001; 26: 1850–7.
11. Wilke HJ, Kettler A, Goetz C, et al. Subsidence resulting from simulated postoperative neck movement: An in vitro investigation with a new cervical fusion cage
. Spine 2000; 25: 2762–70.
12. Cunningham BW, Haggerty CJ, McAfee PC. A quantitative densitometric study investigating the stress-shielding effects of interbody
spinal fusion devices—Emphasis on long term functions in thoroughbred racehorses. Presented at the Annual Meeting of the Orthopaedic Research Society, New Orleans, 1998.
13. Togawa D, Bauer TW, Brantigan JW, et al. Bone graft incorporation in radiographically successful human intervertebral body fusion cages. Spine 2001; 26: 2744–50.
14. Kandziora F, Schollmeier G, Scholz M, et al. Influence of cage
design on interbody
fusion in a sheep cervical spine model. J Neurosurg 2002; 96: 321–32.
15. Ray CD. Threaded titanium cages for lumbar interbody
fusions. Spine 1997; 22: 667–79.
16. Kuslich SD, Danielson G, Dowdle JD, et al. Four-year follow-up results of lumbar spine arthrodesis using the Bagby and Kuslich lumbar fusion cage
. Spine 2000; 25: 2656–62.
17. Boden SD, Zdeblick TA, Sandhu HS, et al. The use of rhBMP-2 in interbody
fusion cages. Definitive evidence of osteoinduction in humans: A preliminary report. Spine 2000; 25: 376–81.
18. Elias WJ, Simmons NE, Kaptain GJ, et al. Complications of posterior lumbar interbody
fusion when using a titanium threaded cage
device. J Neurosurg 2000; 93 (suppl): 52.
19. O’Dowd JD, Lam K, Mulholland RC, et al. BAK cage
: Nottingham results. Presented at the NASS 13th Annual Meeting, October 29, 1998.
20. Heim SE, Abitbol JJ. Complications and strategies for salvage of intervertebral fixation devices. Orthop Clin North Am 2002; 33: 393–8.
21. McAfee PC, Cunningham BW, Lee GA, et al. Revision strategies for salvaging or improving failed cylindrical cages. Spine 1999; 24: 2147–53.
22. Lieberman IH, Willsher PC, Litwin DE, et al. Transperitoneal laparoscopic exposure for lumbar interbody
fusion. Spine 2000; 25: 509–14.
23. Regan JJ, Yuan H, Mcafee PC. Laparoscopic fusion of the lumbar spine: Minimally invasive spine surgery. A prospective multicenter study evaluating open and laparoscopic lumbar fusion. Spine 1999; 24: 402–11.
24. Staelin ST, Zdeblick TA, Mahvi DM. Laparoscopic lumbar spinal fusion: The role of the general surgeon [review]. J Laparoendosc Adv Surg Tech A 2000; 10: 297–304.
25. Zdeblick TA, David SM. A prospective comparison of surgical approach for anterior L4-L5 fusion: Laparoscopic versus mini anterior lumbar interbody
fusion. Spine 2000; 25: 2682–7.
26. Klemme WR, Owens BD, Dhawan A, et al. Lumbar sagittal contour after posterior interbody
fusion: Threaded devices alone versus vertical cages plus posterior instrumentation. Spine 2001; 26: 534–7.
27. Whitecloud TS III, Castro FP Jr, Brinker MR, et al. Degenerative conditions of the lumbar spine treated with intervertebral titanium cages and posterior instrumentation for circumferential fusion. J Spinal Disord 1998; 11: 479–86.
28. Rosenberg WS, Mummaneni PV. Transforaminal lumbar interbody
fusion: Technique, complications, and early results. Neurosurgery 2001; 48: 569–74.
29. Agazzi S, Reverdin A, May D. Posterior lumbar interbody
fusion with cages: An independent review of 71 cases. J Neurosurg 1999; 91 (suppl): 92.
30. Brantigan JW, Steffee AD, Lewis ML, et al. Lumber interbody
fusion using the Brantigan I/F cage
for posterior lumbar interbody
fusion and the variable pedicle screw placement system: Two-year results from a Food and Drug Administration investigational device exemption clinical trial. Spine 2000; 25: 1437–46.
31. Hacker RJ, Cauthen JC, Gilbert TJ, et al. A prospective randomized multicenter clinical evaluation of an anterior cervical fusion cage
. Spine 2000; 25: 2646–54.
32. Profeta G, de Falco R, Ianniciello G, et al. Preliminary experience with an anterior cervical microdiscectomy and interbody
fusion (Novus CT-Ti) in patients with cervical disc disease. Surg Neurol 2000; 53: 417–26.
33. Matge G. Cervical cage
fusion with 5 different implants: 250 cases. Acta Neurochir (Wien) 2002; 144: 539–49.
34. Lenke LG, Bridwell KH. Mesh cages in idiopathic scoliosis in adolescents. Clin Orthop 2002; 394: 98–108.
35. Kandziora F, Pflugmacher R, Kleeman R, et al. Biomechanical analysis of biodegradable polylactic acid interbody
fusion cages. Spine 2002; 27: 1644–51.
36. van Dijk M, Smit TH, Sugihara S, et al. The effect of cages stiffness on the rate of lumbar interbody
fusion: An in vivo model using poly(l-lactic acid) and titanium cages. Spine 2002; 27: 682–8.