Surgical Principle
Depending on the indication, the Wallis implant is placed either subsequent to a conventional decompressive procedure or in isolated fashion through a midline incision. The supraspinous ligament is detached from the two spinous processes of the degenerative lumbar segment and retracted intact with the underlying paravertebral muscles. The interspinous ligament is removed and, if necessary, a decompressive procedure is performed.

Fig 1. The Wallis dynamic stabilisation implant

After determining the size with a trial implant, a spacer is placed in the interspinous space. Two attached tension bands are then secured and tightened around the spinous processes on both sides of the spacer stiffening the segment. The supraspinous ligament is reattached by single sutures through a hole in each spinous process and the wound is closed over a suction drain.

History
Prompted by progress achieved in similar techniques in other joints of the locomotor system, we began studying and developing dynamic stabilisation of lumbar segments. Research in this field had been hindered by the fact that the multiarticular spinal system compensates relatively well for damage to a single segment, regardless of whether such lesions result from a degenerative process or surgical fusion. The stretching of the elements of articular union leads to a force resisting the displacement. The dissipation of kinetic energy in the form of heat is mediated by the viscoelastic properties of the connective tissue (passive damping). This damping phenomenon would, in fact, be quite insufficient to protect the disc if it were not constantly supplemented by a much more effective active damping provided by the reflex contraction of the powerful paravertebral muscles. Although the dynamic equilibrium of the intervertebral articular system is dependent on a combination of muscle activity and tension of the passive elements of union, the active system constantly protects the passive elements, which consequently are never submitted to the limits of their elasticity under normal conditions.

Under these specific mechanical conditions, the intervertebral disc cells that produce the extracellular matrix exhibit normal activity. These cells are, in fact, mechanodependent, as demonstrated by Lotz and Chin [6]. They function normally only under a precise range of mechanical loading. Outside of this range, they initiate apoptosis. When loading is excessive or the active system of damping is deficient, the passive system represented by the disc and intervertebral ligaments can be overloaded and rupture. If these lesions are not excessively severe, or if the lesional process takes place over time analogously to stress fractures, cell activity can repair the damage, as is the case in any connective tissue. However, when the stresses persist, the reparative process can be overwhelmed, and irreversible de-generative lesions develop if the loss of rigidity persists.

Nonetheless, the disc tissue, notably the annulus, has healing capacity, as do all connective tissues. In fact, an indisputable healing process can be observed in the intervertebral disc, with a fibroelastic reaction and neovascularization, at least at the beginning of degenerative lesions.

However, the persistence of excessive mechanical loading leads to the failure of this healing process, similar to that observed in pseudarthrosis of long bones or in meniscal lesions.

Fig 2. A Wallis implant in place

We performed animal studies to test this hypothetical healing capacity of the intervertebral segment under propitious mechanical conditions. Encouraged by these preliminary unpublished results, we carried out biomechanical cadaver studies, mechanically testing various dynamic systems of stabilisation of lumbar intervertebral segments, from 1984 to 1986. The system that we developed and first implanted in 1986 included a titanium interspinous blocker and a woven polyester cord.

The results of an initial observational study were published in 1988 [12, 14]. This was followed by a prospective controlled study from 1988 to 1993 [15]. The results of these studies were quite promising. Subsequently, more than 300 patients were treated for degenerative lesions with the first generation implant, with clinical and radiological follow-up. Assured of the complete innocuousness of the technique by the absence of serious complications in the latter patients over a 10-year period and after careful analysis of the points that could be improved, we developed a second generation implant called the “Wallis” system (Figs. 1, 2), which was marketed in 2002 [13].

Advantages

Technical Advantages

• No bony fixation, as we think it is illusory to hope for durable purchase of bone screws in a dynamic intervertebral construct.
• Due to its shape and its composition in polyether-etherketone (PEEK), the interspinous spacer is 30 times more elastic than the first generation (titanium) spacer to reduce the risk of spinous process stress fractures.
• Flat bands of woven polyester replace the former woven polyester cord to avoid stress concentration on the spinous processes and thus reduce the risk of osteolysis.
• Anatomical angles of the grooves for better fit in the interspinous process space.
• New system of attachment between the spacer and the bands eliminating the torque present in the former system at final tightening.
• The established long-term biocompatibility of PEEK and woven polyester.
• Completely radiolucent materials compatible with MRI.
• Radiodense markers for convenient radiographic follow-up.
• Reduced number of dedicated instruments for rapid placement of the implant.

Clinical Advantages

• Increases stiffness and restores stability to the treated segment relieving low back pain due to degenerative instability.
• Limits but does not eliminate movement in the treated segment (Fig 3).
• Based on finite element analysis modeling the implant’s effect on an uninjured lumbar segment and the intradiscal pressure, Wallis appears not to alter the mechanical behavior of the adjacent segment, possibly limiting the domino effect of degenerative disc disease.
• An interspinous spacer displaces mechanical loading dorsally reducing the load upon the disc and facet joints (by as much as 50% for a spacer 12 mm in thickness [8]).
• With the exception of the interspinous ligament and occasional slight trimming of the spinous processes, preservation of the vertebral and inter vertebral structures.
• No pedicle screws, meaning less invasive procedure with fewer complications.
• Blood loss involving implant placement is negligible.
• Rapid placement reducing operative duration.
• The procedure is fully reversible, leaving all subsequent surgical options open, including fusion, disc replacement, and potential future cell therapeutic interventions [3].

Fig 3. Lateral films showing flexion (left), neutral position (middle), and extension (right). As shown by the lines drawn for the second level of this two-level procedure, extension appears to retain a larger range of motion than flexion in this patient. The radiodense markers of both implants are visible as are the clipping rings added to prevent fraying of the bands at each extremity.

Disadvantages

• The disadvantages of the first generation implant have been identified and taken into account when developing the second generation implant (see above).
• We believe that it is not possible to rigidify all joint elements of the inter vertebral segment with a simple system. In our system of dynamic stabilisation, only flexion/extension and rotation are damped.
• To avoid problems involving screw fixation in a dynamic construct, the Wallis system is not provided with implants for the L5-S1 segment.

Indications
The Wallis mechanical normalization system treats low back pain that accompanies degenerative lesions of grades II, III, and IV (among the five grades in the MRI classification proposed by Pfirrmann et al. [10]) in lumbar segments down to L4-5 in the following indications:

• Voluminous herniated disc in young adults
• Recurrent herniated disc
• Herniated disc accompanying an L5 sacralization transitional anomaly
• Degenerative disc disease at a segment adjacent to fusion
• Modic I degenerative lesions
• Lumbar canal stenosis treated by undercutting (“recalibrage” in French)

Contraindications

• Grade V degenerative lesions in the MRI classification of Pfirrmann
• Spondylolisthesis
• Osteoporosis
• Non-specific low back pain
• Patent constitutional or acquired spinous process insufficiency
• L5-S1
• Litigation
• Patent psychological disorders

Patient’s Informed Consent
It is important for surgeons to inform patients of the inherent complications of the spinal procedures performed under the same anaesthesia as the implant. Regarding the Wallis dynamic stabilisation procedure itself, the surgeon need only provide information on potential complications of anaesthesia and spine surgery, in general (e.g., allergy to anaesthesia, infection, deep venous thrombosis, pulmonary embolism).

It is just as important to inform the patient of the possibility of persistent low back pain due to concomitant degenerative lesions in other discs or facet joints, unless this possibility has been eliminated by methodical diagnostic procedures performed over several days prior to the operation [9, 11].

Surgical Technique
The procedure to insert a Wallis implant is typically associated with minimally invasive unilateral decompression consisting in discectomy, undercutting to enlarge the spinal canal, or both. Only the implant placement will be described here. The decompressive procedures will not be addressed.

Preoperative Planning and Preparation of the Patient

• Plain films
• Bending films
• MRI

Anaesthesia
The operation is performed with the patient under general anaesthesia, but with no precautions specific to the Wallis procedure alone.

Positioning
The patient is placed in the prone position on a frame of foam padding. A neutral position of physiological lumbar lordosis is best to optimize the effect of the implant. All efforts should be made to avoid subsequent lumbar kyphosis.

Surgical Steps

Localization
Carefully locate the segment requiring the implant with an image intensifier.

Skin to Interspinous Space
After a short midline incision, the supraspinous ligament is detached from the two spinous processes at the level involved and retracted laterally without sectioning. After the initial incision, it is recommended to temporarily suture a small sterile surgical drape over the wound edge on both sides to prevent contact between the skin and the implant.

The interspinous ligament is respected. Rough edges of the inferior aspect of the upper spinous process and of the superior aspect of the lower spinous process are trimmed, if necessary, to facilitate insertion of the interspinous spacer.

The bony junction between the laminae and spinous processes may also be trimmed to position the implant as anteriorly as possible and ensure a stable fit against the laminae. Note: If the procedure includes enlargement of a stenotic lumbar canal by resection of the upper portion of the laminae, make sure to preserve sufficient spinous process thickness.

Placement and Securing of the Implant
To preserve physiological lordosis, when hesitating between two implant sizes, the surgeon should choose the smaller implant. Given the prone position of the patient during the operation, substantial primary implant stability is not necessary.

The spacer is positioned in the interspinous space and the bands are passed through the interspinous ligament as close as possible to the instrumented spinous process. On each side, the system is secured by positioning and snapping onto the spacer a clip, through which the band is inserted. The bands are then tightened and a ring is crimped onto each band to avoid fraying after cutting off the excess band.

Reinsertion of the Supraspinous Ligament
The supraspinous ligament is returned to its original position and reinserted onto each spinous process with a single silk passed through a hole made in the spinous process using a Backhaus towel clamp.

Fig 4. Patient with MRI evidence of disc rehydration at follow-up

Final Operative Procedures
To permit proper drainage of possible postoperative epidural bleeding related to a decompressive procedure, the operative wound should be closed over a suction drain on one side of the spacer.

Postoperative Care

• The suction drain is removed the following day in the absence of abnormal discharge.
• External lumbar support is recommended for 1month.
• Patients should refrain from rehabilitation exercises for 1 month.

Results
A prospective single-arm open study of the second generation Wallis implant is underway, in which to date 220 patients (37% women, 63% men; age range 18–82 years; average age 44 years) have been included. Thirty-six percent of the patients presented with degenerative disc disease without disc herniation, 30% with voluminous disc herniation, 18% with canal stenosis, and 16% with recurrent disc herniation.

The mean operating time for implant placement was 19 minutes and the mean time for surgery skin to skin was 79minutes. The average blood loss was 190 cc. Forty-eight percent of the patients received a 12-mm Wallis implant and 38% received a 10-mm Wallis implant.

The most frequently operated level was L4-5 (92% of the patients).

Among the 220 patients, there were six (2.72%) postoperative complications. Three patients developed deep infection leading to removal of the implants. One patient with psychological problems had the implant removed by another surgeon 10 months after the intervention. Two implants were replaced by a second Wallis implant, one after immediate recurrence of herniated disc and the other after implant displacement. Although these were considered as adverse events, they confirmed the straightforwardness of Wallis implant replacement and that all options remained for other surgical solutions.

Preliminary, 12-month results have been obtained for 52 patients involved in the ongoing multicenter international study.

Preoperative evaluation of the functional and pain status of the patients was performed using the Japanese Orthopedic Association assessment of lumbar pain management (JOA score) [16], a visual analog scale for lumbar pain (VAS), and two quality-of-life questionnaires, the first version of the medical outcomes score short form 36 (SF-36) [4] and the Oswestry low back pain disability questionnaire [2].

In the preoperative period, 65% of the patients had pain scores between 70 and 100 on the VAS and 90% of the patients had pain scores between 0 and 30 1 year after surgery.

The result of the straight leg-raising test was negative in 92%of the patients. Sensory and motor disturbances were present in 75% and 61.5% of the patients before surgery and in 10% and 4% of the patients after 1 year, respectively.

The therapeutic success as assessed by Odom’s criterion at 1-year follow-up found 89.6% of the patients with excellent or good clinical results, 6.25%with satisfactory results, and 4.15% with poor results.

Critical Evaluations
The Wallis system is only applicable above L5 and does not include L5-S1. Although much work has been done and solutions have been found, none have been sufficiently satisfactory.

Consequently, we have intentionally limited the application not to include the lumbosacral junction. A bilateral approach is necessary for the moment in order to secure the implant with the bands, but work is in progress to develop an implant that can be inserted through a unilateral approach. There is also a small risk of recurrent disc herniation due to the reduced, but persistent mobility of the segment.

Fig 5. Another patient with MRI evidence of disc rehydration at follow-up

Next to these points of criticism, the present system has many substantial advantages. It is technically simple and straightforward, with a short learning curve and reduced perioperative and postoperative morbidity. The early mobilization and rehabilitation of operated patients are also appreciable.

One of the major incentives prompting the development of dynamic stabilisation was its potential protective effect against accelerated changes in adjacent discs seen after fusion. Years of follow-up will be necessary to establish whether this theoretical advantage is actually achieved. Furthermore, although it is still too early to determine whether the treated disc tissue actually heals or undergoes scarring, there is MRI evidence of disc rehydration at follow-up.

Many initially black discs are coming back with normal, white signal. Exam-ples are shown in Figs. 4 and 5.

Among the notions to retain concerning dynamic lumbar stabilisation, foremost are the good clinical results being observed. Follow-up evidence has convinced us that, to date, the efficacy of the new implant is at least as good as that of fusion.

Wallis is an operative alternative for many young patients with degenerative disc disease who suffer, but for whom fusion, or even a disc prosthesis, might seem too radical. Surgery rarely affords definitive solutions, and spinal surgery is no different. If the threshold of surgery is low and if a surgical procedure leaves other options open, a step-by-step strategy to treat low back pain without compromising future solutions should be considered. Through the Internet, more and more patients are aware of the advent of biological methods of treating degenerative discs with the patients’ own stem cells or fibroblasts [3]. This perspective is at our doorstep.

Stem cell therapy has already begun for tendon and ligament lesions [5]. For patients and surgeons, Wallis is a failsafe solution because the procedure is completely reversible. If relief from low back pain is not achieved or if, after years of relief, the degenerative process regains the upper hand, surgeons will still have all the original options to treat their patient. Consequently, this method should rapidly assume a specific role along with total disc prostheses in the new stepwise surgical strategy for early forms of degenerative disc disease.

References

1. Ebara S, Harada T, Hosono N, et al (1992) Intraoperative measurement of lumbar spinal instability. Spine 17:44–50
2. Fairbank JCT, Mbaot JC, Dvies JBD, Oebrien JP (1980) The Oswestry low back pain disability questionnaire. Physiotherapy 66:271–273
3. Ganey TM, Meisel HJ (2002) A potential role for cell-based therapeutics in the treatment of intervertebral disc herniation. Eur Spine J 11(suppl 2): S206-S214
4. Glassman SD, Minkow RE, Dimar JR, Puno RM, Raque GH, Johnson JR (1998) Effect of prior discectomy on outcome of lumbar fusion: a prospective analysis using the SF-36 measure. J Spinal Disord 11:383–388
5. Hildebrand KA, Jia F, Woo SL (2002) Response of donor and recipient cells after transplantation of cells to the ligament and tendon. Microsc Res Tech 58:34–38
6. Lotz JC, Chin JR (2000) Intervertebral disc cell death is dependent on the magnitude and duration of spinal loading. Spine 25:1477–1483
7. Mimura M, Panjabi M, Oxland TR, et al (1994) Disc degeneration affects the multidirectional flexibility of the lumbar spine. Spine 19:1371–1380
8. Minns RJ, WalshWK (1997) Preliminary design and experimental studies of a novel soft implant for correcting sagittal plane instability in the lumbar spine. Spine 22:1819–1825
9. Pang WW, Mok MS, Lin ML, Chang DP, Hwang MH (1998) Application of spinal pain mapping in the diagnosis of low back pain-analysis of 104 cases. Acta Anaesthesiol Sin 36:71–74
10. Pfirrmann CWA, Metzdorf A, Zanetti M, Hodler J, Boos N (2001) Magnetic resonance classification of lumbar intervertebral disc degeneration. Spine 26:4873–4878
11. Schwarzer AC, Wang SC, O’Driscoll D, Harrington T, Bogduk N, Laurent R (1995) The ability of computed tomography to identify a painful zygapophysial joint in patients with chronic low back pain. Spine 20:907–912
12. Sénégas J (1991) La ligamentoplastie intervert´ebrale, alternative `a l’arthrodése dans le traitement des instabilitiés dégénératives. Acta Ortop Belg 57(suppl 1):221–226 13. Sénégas J (2002) Mechanical supplementation by non-rigid fixation in degenerative intervertebral lumbar segments: the Wallis system. Eur Spine J 11(suppl 2): S164–S169
13. Sénégas J, Etchevers JP, Baulny D, Grenier F (1988) Widening of the lumbar vertebral canal as an alternative to laminectomy, in the treatment of lumbar stenosis. Fr J OrthopSurg 2:93–99
14. Sénégas J, Vital JM, Gu´erin J, Bernard P,M’BarekM, Loreiro M, Bouvet R (1995) Stabilisation lombaire souple. GIEDA: instabilit´es vert´ebrales lombaires. Expansion Scientifique Fran¸caise, Paris, pp 122–132
15. Yorimitsu E, Chiba K, Toyama Y, Hirabayashi K (2001) Long-term outcomes of standard discectomy for lumbar disc herniation. A follow-up study of more than 10 years. Spine 26:652–657

previous - return to list  - next