Lumbar Interbody Fusion with Titanium Cages: Five Year Follow-Up

Toussaint LeClercq, MD; John J. Kruse, DMD, MD; Deepak Awasthi, MD

Department of Neurosurgery; Louisiana State University Medical Center; New Orleans, Louisiana

The first 20 cases of the 100 patients treated with lumbar interbody fusion with threaded titanium cages are analyzed in this report. The indications for this procedure included combinations of low back pain, loss of disc height and osteophytes, recurrent herniated lumbar disc, spondylolisthesis (grade I), disrupted annulus, segmental instability and herniated lumbar disc. Two cages packed with autogenous bone were implanted per level after the standard preparation of the intervertebral disc space. Follow-up was done at 3, 6 and 12 months and yearly thereafter. The patients' clinical exam and symptoms were recorded. Lumbar spine x-rays and, at times, lumbar CT scan was performed during the follow-up period.

The first 20 cases (14 males and 6 females) have been followed for 5-6 years (average 68 months). Twenty-two levels were operated on (9 at l4-5 and 13 at L5-S1). All patients were fused at one year follow-up as determined by spine x-rays and dynamic views. The disc height was maintained at 11.7 mm in ten levels. Ten levels went from a mean of 6.8 mm pre-operatively to 11.1 mm postoperatively. Two levels collapsed postoperatively (in the same patient). Pre-operatively, 16 patients were disabled and 4 were working on light duty. At 68 months follow-up, 13 patients were working (2 on light duty); 2 were disabled, 3 were lost to follow-up and 2 were deceased. Thus, in this small series with a significant follow-up, the posterior lumbar interbody fusion procedure with titanium cages offers a safe and effective means of clinical and functional improvement.

[Introduction | Materials & Methods | Results | Discussion | References]

Posterior lumbar interbody fusion (PLIF) using titanium cages has gained momentum in the recent years after approval of the cages by the FDA (United States Federal Drug Administration) for lumbar interbody insertion. The indications for this procedure remain controversial and include mechanical low back pain, degenerative disc disease, recurrent disc herniation, spondylolisthesis (grade I).(6,7,9,10,11,13,15) In degenerative disc disease or following surgical discectomy, segmental stenosis occurs due to a combination of disc herniation, posterior osteophyte formation, facet overriding and hypertrophy and ligamentum flavum hypertrophy. (7) All these factors combine to compromise the nerve root as it exits through the intervertebral foramen resulting in recurrent radiculopathy. In such cases, PLIF using cages can open up the intervertebral foramen by maintaining or restoring disc height.

On the other hand, there is considerable controversy surrounding the effectiveness of lumbar fusion for the treatment of low back pain. Among the different surgical approaches available, with or without instrumentation, none can be considered the "gold standard". Moreover, most reports deal with short-term follow-up when, in fact, results on low back surgery are best analyzed on a long-term basis.(4) The technique and results of PLIF were first reported by Cloward. (2,3) It did not, however, gain widespread acceptance due to technical difficulties. Insertion of a rectangular bone graft makes it difficult to avoid nerve root trauma resulting in radicular deficit or irritation and dural tears. Non-union or bone graft extrusion were reported as well as bone graft collapse with subsequent segmental stenosis. (5,9,17) In the past few years, rigid cages housing autogenous bone have become increasingly ppoular for the purpose of interbody fusion. Their ease of insertion has decreased the technical difficulties of the earlier Cloward approach and makes the procedure more reproducible. The rigidity of the cages also allows for the preservation of the disc space. In order to analyze our experience with this procedure we selected the first 20 cases of the 100 done at LSUMC. These cases have the longest follow-up: average of 68 months.

Materials and Methods:
Surgical Technique:
The procedure is done under general anesthesia with the patient positioned in knee-chest to decrease epidural bleeding. Fluoroscopic control is used to mark the level(s) to be fused as well as permits the assessment of the physiologic curve of the spine, disc height and accurate placement of the cages. After identification of the appropriate level, a standard bilateral laminectomy is performed with preservation of the midline structure (spinous process and ligaments). One-third to half of the articulating facets have to be resected bilaterally in order to facilitate the insertion of the cages with minimal morbidity to the nerve roots. The operative microscope or loupe magnification and microdissection techniques are used to fully identify the nerve roots in the epidural space. The disc material is then removed bilaterally and a vertebral drill (with the appropriate depth) is used to obtain satisfactory purchase into the end plates. Using dilators of increasing diameters, the vertebral bodies are distracted to obtain a satisfactory disc height. Threads are then cut into the opposing vertebral surfaces using a bone tap. The space is then ready to receive the threaded cage (of the appropriate diameter and height) which is screwed to lie beyond the posterior vertebral cortices. The cage is then packed with autogenous bone and the incision is closed in layers.

The threaded fusion cages (Figure 1)are hollow alloy cylinders which are perforated on 70% of their cylindrical walls.

Figure 1: Photograph of the two types of cages most commonly used in the US. On the left is the Ray titanium threaded cage. The cage in the model spine on the right is the BAK cage.

Product compression tests have shown that each cage can be loaded up to 900 newtons for up to 10 million cycles without failures. In addition, the cages are MRI compatible.The cases analyzed had the Ray cages used for the PLIF. The Ray threaded fusion cage is a registered mark of Surgical Dynamics, a subsidiary of united Surgical Corporation, Norwalk, Connecticut. In addition, BAK cages (SpineTech) are also available for PLIF as well as the anterior lumbar interbody fusion (ALIF) procedure.

Between June 1992 and May 1993, 20 patients were operated at 22 levels for lumbar interbody fusion. All patients presented with symptomatic lumbar disc disease associated with low back and radicular pain. The low back pain had characteristics of mechanical pain (increasing with activity). Seven patients had previous surgical discectomy at the same level and suffered with recurrent disc herniation and epidural scarring as evidenced by a lumbar MRI (with and without contrast). Three patients presented with degenerative Grade I spondylolisthesis.

The age of the patients varied from 26 to 65 with a mean of 38.8 years. There were 14 males and 6 females. The pre-operative neurological findings included: motor deficit (2 patients); sensory deficit (4 patients); positive straight leg raising (16 patients); decreased or abolished deep tendon reflex (14 patients); palpable muscle spasm (10 patients); limitation of back movements (19 patients). The symptoms were present for a mean of 18.5 months (5-72 months). The neuroradiologic work-up consisted of a lumbar MRI scan (20 cases), CT scan (17 cases), CT myelogram (1 case) and CT discograms (8 cases). The disc height was measured pre-, intra- and post-operatively as well as the last follow-up for each case. The surgery was performed at the L4-5 level in 9 cases and at the L5-S1 level in 13 cases.

Follow-up was performed at 6 weeks; 3, 6, 12 and 24 months. Yearly follow-up was done thereafter whenever feasible. The Prolo Scale (13) was used to assess the patients' clinical and finctional status pre- and post-operatively and at follow-up. Criteria for fusion included: no halo of bony reabsorption around the cage; maintained bony density within the cage and no visible movement on flexion-extension. The sacrolumbar junction is best studied with Ferguson views (AP view of the spine with 30-35 degrees cephalad inclination of the X-ray tube). In some cases, thin cut CT scans were used to demonstrate the presence of bone inside the cage but postoperative CT or MRI scans were not done routinely for economical reasons.

The Prolo Scale (13) was used to assess the patients pre- and post-operatively. The results can be found in Table 1 below:

E1 Complete invalid
E2 No gainful occupation 16 1 1 1
E3 Able to work, but not at previous occupation 4 3 2
E4 Working at previous occupation on a part time or limited status 2
E5 Able to work at previous occupation with no restrictions of any kind 15 13 11
F1 Total incapacity (or worse than before the operation)
F2 Mild to moderate level of back pain and/or sciatica 19 1 1 1
F3 Low level of pain and able to perform all activities except sports 2
F4 No pain, but patient has had one or more recurrence(s) of back pain and/or sciatica 3 3 4
F5 Complete recovery, no recurrent episodes of back pain, able to perform all previous sports activities 1 13 13 10

Thirteen of 20 patients (65%) made a complete recovery with no recurrent episodes of pain at 1 and 3 years. At 5 year follow-up 10 of 20 patients (50%) remained without recurrence and an additional 4 had no pain, but had one or more recurrence(s) of back pain. Pre-operatively, 16 patients were disabled and 4 were working on light duty. The pre-operative low grades of the patients are due to the fact that, except in very rare cases, the lead author will not perform lumbar fusion operations on patients who can perform their work. At 68 months follow-up, 13 patients were working (2 on light duty), 2 were disabled, 3 were lost to follow-up and 2 were deceased. Overall, the combined Prolo Scale improved from 4 pre-operatively to 9.2 postoperatively.
There were no cases of infection, radicular deficit or CSF leak postoperatively. One patient was re-operated to reposition a cage producing radicular pain secondary to improper placement. Repositioning of the cage presented no difficulty in technique and resulted in radicular pain relief post-operatively. At the time of the second operation (6 weeks after the original procedure) for cage repositioning, the bone inside the cage was found to be well vascularized. Growth of bone between the cage and the verebral bodies was present. The patient went on to a satisfactory fusion.

Using the criteria described in the methods section, satisfactory fusion was obtained at all 22 levels at one year follow-up. The results were maintained at 5 years using similar criteria in all cases available for follow-up. There were no cases of late breakage or displacement of the cages. In some cases the bony fusion could be demonstrated outside the cage (Figure 2).

Figure 2: Sagittal CT reconstruction done approximately 3 years after cage placement showing calcification outside the cages.

Disc Height:
At 10 levels, the disc space was increased from a pre-operative height of 6.8 mm (average) to 11.1 mm (average) post-operatively- see Figure 3.

Figure 3: A: Pre-operative lateral lumbosacral x-ray showing collapse of the L5-S1 disc space. B: Intraoperative x-ray of teh same patient showing increase in the disc height at L5-S1 after cage placement. C: Long term (approximately 2 years) postoperative x-ray of the same patient showing continued maintenance of the increased disc space height at the L5-S1 level (site of cage insertion).

Ten levels were considered as "maintained" when the pre-operative space was considered satisfactory and, thus, no effort was made to increase it during surgery. The average value for these ten levels was 11.7 mm. In one patient operated operated on at 2 levels (L4-5 and L5-S1), both spaces were found to be collapsed at the 3 month follow-up x-ray. The cages were found embedded into cancellous bone of the vertebral bodies. Despite this collapse, both levels went on to fuse and the patient progressed satisfactorily and returned to his original work.

Posterior lumbar interbody fusion has been performed for over 50 years. (2,3). Technical difficulties have precluded wide acceptance of the technique. Failures in the past have included collapse of the bone graft in the disc space, reabsorption of the graft and pseudo-arthrosis. Extrusion of the graft has resulted in nerve root compression with radicular deficit. In addition, there can be complications of harvesting a bone graft from the hip.(10) The titanium cage has attempted to solve some of these technical difficulties by providing a rigid construct and safer instrumentation. Thus, the PLIF procedure utilizing cages has gained wider acceptance in recent years.
The place for lumbar fusion in the treatment of low back pain has been a matter of debate. The role of the intervertebral disc as a pain generator is not widely recognized and many surgeons emphasize the nerve root compression component as the major factor to be addressed by surgical treatment. Nevertheless, the intervertebral disc as a pain generator is slowly gaining wider acceptance (8,12,16) and many people advocate removal of disc with or without herniation and nerve root compression (11,15).

Rationale for Fusion with Cages:
When deciding to perform a lumbar fusion, it seems logical to fuse the anterior column of the spine which bears 80% of the weight bearing forces. The adbavtages of interbody fusion include direct removal of the dysfunctional disc and preservation or restoration of the disc height. Maintenance of the disc height, especially posterior aspect is important to achieve significant increase in the neuroforamen volume. (1) Titanium interbody devices can better preserve the disc height than bone (autogenous or heterogenous) as shown by sheep studies.(14) In interbody fusion done bone alone, Dennis et al showed a loss of disc height in all patients undergoing anterior lumbar interbody fusion.(5) In the series presented, preservation of disc space, whether it was maintained or restored, was obtained immediately postoperatively and the results were preserved at a long-term follow-up (68 months).

Technical Points:
We have elected to do surgery in the knee-chest position. Though some surgeons are critical of this position, a postoperative flat back syndrome was not observed in any of the cases. Indeed, in some of the cases, we were able to improve the lordotic curve of the lumbar spine, especially in patients with previous back surgery.

The extent of the laminectomy necessary to insert the cages depends upon the surgeons' preference and comfort level. Many surgeons prefer to perform a total laminectomy with removal of the spinous process and interspinous ligament. Although necessary in select cases, we prefer to preserve the midline structures (spinous process and inter/supraspinous ligaments) because these structures may play an important role in the stability of the posterior column. Cages can be inserted by removing about 5 mm of the laminae above and below the intervertebral disc level to be fused. In addition, it is usually necessary to perform a partial (usually 1/2, sometimes 1/3) facetectomy in order to facilitate cage insertion. This can be done with a hammer and a 5mm osteotome and the bone can be saved to fill the cages. In some cases, a high speed drill may be helpful to assure a more comfortable acceptance for the retractor and cages.

Clinical Results and Future Studies:
Results must be judged on radiological and ,especially, clinical criteria- patients' satisfaction. The immediate rigidity obtained with cages results in less postoperative pain, shorter hospitalization and overall reduced cost.(6) Our results confirm these data. Results on low back surgery are best analyzed on a long-term basis.(4) When asked if they would have the surgery again, 12 out of 15 patients available at 5 year follow-up in our study responded affirmatively and estimated the improvement obtained with surgery at around 85%.

Ideally, randomized studies should be performed to assess which technique of interbody fusion or if interbody fusion is best. Unfortunately, at the present time, we have no "gold standard" for spinal fusion. The need for patient informed consent makes randomization difficult since patients choose what appears to be least traumatic and easier option in their mind. In the USA the fact that postero-lateral fusion associated with pedicle screws involves the use of instrumentaion which is not FDA approved further complicates the availablity of randomized controlled studies.

Lumbar fusion with titanium cages affords the spine surgeon with a safe and reliable technique to obtain spinal fusion. Whether the goal of maintaining or improving the disc height will improve the clinical results remains unproven at this time. Patients with satisfactory fusion and collapsed disc space do well. Larger series comparing satisfactory fusions with maintained or collapsed disc spaces need to be established to confirm that disc height maintenance is indeed beneficial to the patients.


  1. Chen D, Fay LA, Lok J, Yuan P, Edwards WT, Yuan HA: Increasing neuroforaminal volume by anterior interbody distraction in degenerative lumbar spine. Spine 20:74-79, 1995.
  2. Cloward RB: New treatment of ruptured intervertebral disc. Presented at the Annual Meeting of the Hawaii Territorial Medical Association, May 1945.
  3. Cloward RB: The treatment of ruptured intervertebral discs by vertebral body fusion: Indications, operative technique, after care. J Neurosurg 10:154-168, 1953.
  4. Davis RA: A long term outcome analysis of 984 surgically treated herniated lumbar discs. Neurosurgery 80:415-421, 1994.
  5. Dennis S, Watkins R, Landaker S: Comparison of disc heights after lumbar interbody fusion. Spine 14:876-878, 1989.
  6. Hacker JR: Comparison of interbody fusion approaches for disabling low back pain. Spine 22:660-666, 1997.
  7. Inoue S, Watanabe T, Hirose A: Anterior discectomy and interbody fusion for lumbar disc herniation. A review of 350 cases. Clin Orthop 183:22-31, 1984.
  8. Kuslich S, Ulstrom C, Michael CJ: The tissue origin of low back pain and sciatica: A report of pain response to tissue stimulation during operations on the lumbar spine using local anesthesia. Orthop Clin North Am 22;181-187, 1991.
  9. Lin PM: Posterior Lumbar Interbody Fusion. Charles C. Thomas Ed: Springfield, Illinois, 1981, p 308.
  10. Matage G: Anterior Interbody Fusion with the BAK cage in cervical spondylosis. Acta Neurochir (Wien) 140:1-8, 1998.
  11. Newman MH, Grinstead GL: Anterior lumbar interbody fusion for internal disc disruption. Spine 17: 831-833, 1992.
  12. Olmarker K, Blomquist J, Stromberg J, Nannmark U, Thomasen P, Rydevik B: Inflammatogenic properties of nucleus pulposus. Spine 20:665-669, 1995.
  13. Prolo DJ, Oklund SA, Butcher M: Toward uniformity in evaluating results of lumbar spine operations: A paradigm applied to posterior lumbar interbody fusions. Spine 11:601-606, 1986.
  14. Sandhu HS, Turner S, Kabo JM, Kanim LE, Liu D, Nourparvar A, Delamarter RB, Dawson EG: Distractive properties of a threaded interbody fusion device. An in vivo model. Spine 21:1201-1210, 1996.
  15. Schechler NA, France MP, Lee CK: Disc disease update- Painful internal disc derangements of the lumbosacral spine: discogenic diagnosis and treatment by posterior lumbar interbody fusion. Orthopaedics 14:447-451, 1991.
  16. Schwarzer AC, Aprill CN, Derby R, Fortin J, Kine G Bogduk N: The prevalence and clinical features of internal disc disruption in patients with chronic low back pain. Spine 20:1878-1883, 1995.
  17. Stonecipher T, Wright S: Posterior lumbar interbody fusion with facet-screw fixation. Spine 14:468-471, 1989.