Is the Spondylolisthesis Responsible for Symptoms?

In evaluating a patient with mechanical, nonradicular back pain without neurologic findings noted to have a spondylolisthesis, an initial question might be whether the slip is responsible for symptoms or is simply an incidental radiographic finding. Most patients with spondylolisthesis are asymptomatic and progression of clinical symptoms does not correlate with progression of the slip.⁴ The clinical significance of “instability,” characterized by change in listhesis demonstrable on flexion and extension lumbar lateral radiographs, is unknown and not clearly correlated with symptoms. Although patients with spondylolisthesis do experience more daily back pain, they do not report greater disability than individuals without a slip.⁵ The most common clinical entity associated with spondylolisthesis is symptomatic spinal stenosis (i.e., pseudoclaudication) in older patients with a degenerative slip.

What Is the Natural History of Axial Back Pain without Neurologic Findings in Patients with Spondylolisthesis?

Irrespective of pathogenesis, the natural history of axial low back pain in patients with spondylolisthesis is favorable. Most patients do well with conservative care.² A prospective cohort study demonstrated that 76% of patients with degenerative spondylolisthesis and no neurologic deficits followed for more than 10 years remained without neurologic findings.⁴ Patients with neurologic symptoms, however, including pseudoclaudication or cauda equina symptoms, deteriorated without surgery. A meta-analysis of nonoperative treatment results in children and young adults with spondylolysis and isthmic grade 1 spondylolisthesis reported 83.9% of patients had a successful clinical outcome at 1 year.⁶ Interestingly, a successful treatment outcome was not dependent on healing of the spondylolytic lesion.

What Is the Optimal Medical/Interventional Treatment for Axial Back Pain without Neurologic Findings in a Patient with Spondylolisthesis?

A recent evidence-based review of medical and interventional treatment for degenerative spondylolisthesis found no high-quality studies to adequately address the question of whether nonoperative treatment improves the generally favorable natural history of this condition when neurologic symptoms, particularly pseudoclaudication, are absent.² Lacking data, treatment of chronic axial back pain in patients with spondylolisthesis should be based on well-established, consensus treatment recommendations for chronic back pain without a slip. Exercise, both general aerobic conditioning and spine specific, is the cornerstone of treatment. Recent systematic reviews of exercise in nonspecific chronic low back pain have confirmed the efficacy of different approaches, including lumbar extensor strengthening, lumbar stabilization, and the McKenzie method.^7–9 An older, small retrospective study comparing flexion and extension exercise programs in patients with spondylolisthesis and back pain found significantly better functional outcome and pain relief at 3 years in the flexion exercise group.¹⁰ A structured dynamic lumbar stabilization exercise program was more effective than usual care in reducing pain intensity and functional disability in a small randomized trial in younger patients with predominantly isthmic spondylolisthesis.¹¹ Multiple studies have demonstrated that general aerobic conditioning is helpful in patients with nonspecific low back pain, but this approach has not been studied specifically in patients with spondylolisthesis.^12–14

In addition to active exercise, age-appropriate symptomatic management with nonopioid analgesics or nonsteroidal anti-inflammatory drugs (NSAIDs) is indicated, as in patients with axial back pain without spondylolisthesis. In a 45-year-old patient without gastrointestinal, cardiovascular, or renal comorbidity, either acetaminophen or an NSAID is a reasonable choice. Muscle relaxants have not been show to have benefit for patients with chronic back pain.

In most patients with chronic nonradicular low back pain and spondylolisthesis, the anatomic source of the pain remains ambiguous. A recent evidence-based systematic review strongly recommended against provocative discography as a diagnostic procedure in such patients, primarily because of a high rate of false positive responses.¹⁵ This study found no evidence that the use of provocative discography to select patients for fusion improved clinical outcomes. Similarly, the same analysis concluded there was “insufficient evidence” to evaluate the utility or validity of medial branch blocks for identification of facet-mediated pain in patients with axial low back pain. However, a recent randomized, double-blind, small controlled trial found that lumbar facet joint neurotomy using radiofrequency current in patients with chronic low back pain was at least statistically significantly superior to a “sham” procedure in improving pain and function.¹⁶ Interestingly, in this trial, only 15% of screened low back pain patients met the stringent diagnostic criteria for facet-mediated pain, suggesting that the facet joint is not commonly the source of chronic low back pain.

There is modest evidence that massage, manipulation, and acupuncture may be helpful adjunctive treatments in management of chronic nonspecific low back pain, but they have not been studied specifically in patients with spondylolisthesis.^17–19 Similarly, a systematic review of the literature found no evidence that ancillary treatments, including bracing, traction, and electrical stimulation, are effective in degenerative spondylolisthesis.²

Summary

The best evidence available and extensive clinical experience suggest that medical management of a 45-year-old patient with mechanical low back pain and L4-5 spondylolisthesis without neurologic findings should include education of the patient about the favorable natural history of the condition and the safety and importance of active exercise, including aerobic activity. A customized exercise program developed by a physical therapist emphasizing a stabilization approach should complement the general conditioning program. Symptomatic management with an NSAID or acetaminophen is indicated with appropriate precautions. Adjunctive complementary therapies may assist in pain management.

Establish the Diagnosis

In deciding treatment options the key is to establish the diagnosis. In a patient with low back pain in the absence of neurologic symptoms, this is often the most difficult and important piece of information. The causes of back pain and the workup are addressed in other chapters. The AANS Guidelines on Surgery for Axial Back Pain indicate that surgery for axial back pain may be an option in those patients with a clear symptomatic level or levels who have failed an appropriate course of active nonoperative treatment.¹¹ Surgical treatment for significant instability in the spine is less controversial, though there is no uniform definition of instability.¹²

In the context of this case the establishment of instability may alter the treatment options selected. Biomechanical studies of stand-alone anterior interbody constructs show that the absence of an intact posterior tension band makes these implants much less stable and that supplemental fixation may be indicated.¹³ In the setting of isthmic spondylolisthesis ventral stand-alone constructs have been used without supplemental fixation in low-grade slips with reasonable results.^14,15 In degenerative spondylolisthesis there has been little written.¹⁶

Disc collapse may also be an important factor in choosing an approach. Stand-alone constructs rely on tension of the fibers of the disc anulus for stability.^17,18 Discs with collapse have shortened fibers and may be more stable. In clinical terms post hoc analysis of patients in FDA trials of a tapered threaded interbody device showed improved outcomes with increasing loss of disc height preoperatively.¹⁹

As a generalization ventral approaches are an option alone or as part of a combined approach in any patient in whom fusion is being considered.

Why a Ventral Approach?

There are a number of biomechanical and clinical reasons why ventral approaches have advantages over other approaches. Historically these approaches were considered highly morbid, but with technical advances they have relatively low morbidity.

In the lumbar spine, the majority of the axial load passes through the disc space. Bone forms better under compressive loads. As a result interbody constructs place the fusion mass in the area of highest load and under the best loading conditions for fusion. Surface area for fusion is also larger in interbody constructs. Predominant loads in the intertransverse region of the spine are tension or shear and the area for fusion is smaller. As a result of the wide anular opening in the ventral approach, disc removal and endplate preparation are easier and more complete than with posterior interbody techniques.²⁰

Fusion rates with anterior interbody fusion are high. In prospective studies fusion rates in the 90% range are common.^21–24 In one large study the radiographic rate of fusion on reconstructed CT imaging was 100%.²² In comparative studies fusion rates with ventral stand-alone constructs are equal to or better than with dorsal fusion or dorsal interbody constructs.^2,14,25 In conjunction with dorsal instrumentation, anterior interbody fusions probably have the most consistently high fusion rates and, in comparison studies against dorsal and dorsal interbody constructs, have shown equivalent or better fusion rates and equivalent or better clinical outcomes.^26–28

The ventral approach in the lumbar spine is technically demanding but with experience can be done quickly and safely.^1,29–31 Vascular injury rates range from 0% to 11% and decrease as experience increases.^29,31,32 Ureteric and enteric injuries are rare.^1,33,34 Retrograde ejaculation rates have been highly variable, but in large series from experienced centers this rate can be reduced to the 1% to 3% range. Shorter hospital stays are reported for ventral-only procedures when compared to open dorsal fusions. Blood loss and operative time are also lower. Though the use of an access surgeon is common, at least two studies have shown that operative times and complications are not decreased through use of an access surgeon.^35,36 Surgical volume and experience with the approach may be more important than the surgical background of the surgeon doing the approach.

Numerous studies have shown that dorsal open surgical approaches lead to histochemical radiographic and pathologic changes to the dorsal musculature.^37–43 Muscle strength is compromised.³⁷ Placement of pedicle screws leads to increased loads on the suprajacent facet.⁴⁴ This constellation of pathology has been termed fusion disease by Zdeblick. The anterior perirectus approach and the minimal access lateral muscle splitting approach avoid these problems. Some authors have suggested that anterior interbody fusion decreases the risk of adjacent segment degeneration as a result.^45–48

Not all patients or pathologic processes are amenable to a ventral approach and in some cases dorsal options alone or combined approaches are a better option. There is limited ability to directly decompress the spinal canal through the ventral approach. Ventral compression by bulging or herniated discs can be dealt with through ventral-only approaches. Dorsal pathology, such as facet or ligamentous hypertrophy, cannot be directly addressed. Indirect decompression is possible. Restoration of disc height can increase foraminal height and volume, thus indirectly decompressing foraminal stenosis.^49–53 Reduction of a spondylolisthesis will also improve foraminal stenosis and in the case of a degenerative spondylolisthesis will also improve central canal diameter.^49,54 In this case, in which no neurogenic symptoms are present, decompression is not a major goal of surgery.

To perform the discectomy and place an interbody graft, it is necessary to have linear access to the disc space. In patients with high slip angles it may not be possible to access the disc; therefore, it may not be possible to do an anterior interbody fusion. When considering a ventral approach, a standing lateral radiograph with the pelvic symphysis visible on the film will allow assessment of the trajectory. As seen in Figure 224-1A, a line drawn along the superior endplate of S1 passes above the symphysis and a ventral approach is possible. In Figure 224-1B the line goes below the symphysis and a ventral approach would not be possible. In general the L4-5 level is more horizontal, so in this case it is unlikely that a ventral approach would not be possible due to approach angle.

FIGURE 224-1 A, Upright lateral radiograph. A line parallel to the endplate of S1 is above the symphysis, and in this case a ventral approach is possible. B, Lateral upright radiograph showing a line along the superior endplate of S1 passing dorsal to the pelvic symphysis. In this patient a ventral approach to L5-S1 is not possible.

Although the risk of retrograde ejaculation is low, it can lead to infertility if it occurs. For this reason, this complication should be discussed with a male who is still considering having children. In some cases presurgery sperm banking may be considered or the patient may elect to not have a ventral approach.

The bifurcation of the inferior vena cava to the common iliac veins generally occurs at or near the L4-5 disc. It is generally visible on the preoperative MRI,⁵⁵ although some advocate a preoperative CT angiogram to assess its position.⁵⁶ It is generally possible to mobilize the vessels off the disc, but in rare cases it is impossible and a ventral approach must be aborted. Extensive vascular calcification and aneurysm of the aorta and common iliac vessels are also relative contraindications to a ventral approach to the lumbar spine due to the possibility of compromising the circulation of the lower extremities. In patients who have had previous retroperitoneal surgery, it is very difficult to access and mobilize the vessels; a ventral approach should be avoided.

In some patients, a minimal access lateral approach allows ventral access to the L4-5 level without mobilizing the vessels. This approach is limited by the height of the iliac wing and there is a risk to the lumbosacral plexus even with EMG monitoring.^8,9

What Conditions Are Amenable to a Ventral Approach?

The ventral approach can be used for most lumbar pathologies. The treatment of intractable low back pain with disc degeneration has been the main indication for stand-alone interbody fusion and has had good short- and long-term results.^{21–23,47,57} Other authors have reported much poorer results.^58–60 The use in both isthmic and degenerative spondylolisthesis has also been established.^{14–16,27,28,61} Anterior interbody fusion has been used in the treatment of herniated discs in those patients in whom fusion is indicated.^62,63

Interbody Fusion Options

Interbody fusions typically involve a structural component to increase and maintain disc height and a biological component to ensure fusion. The early descriptions of anterior interbody fusion used iliac crest bone graft as a tricortical strut.⁶³ Iliac crest bone provides structural support and is osteogenic. Allograft femoral rings have also been widely and successfully used.^{21,45,47,64,65} They provide a strong strut and some osteoconductive properties. Cancellous autograft is typically placed inside the ring, although recent studies have used recombinant human bone morphogenetic protein-2 (rhBMP-2).^66,67 Studies with BMP have shown a higher failure rate when used as a stand alone.^67,68

Numerous interbody implants have been developed using multiple materials. The intitial implants were threaded interbody devices, typically metallic.^20,21,24 Threaded cortical allograft dowels were used in the past, although they are no longer widely used.²² Impacted metallic cages were developed to provide a broader footprint to lessen subsidence.⁶⁹ Metallic interbody implants make imaging to assess fusion difficult and have a modulus of elasticity much higher than cancellous bone. As a result, carbon fiber⁷⁰ and PEEK⁷¹ implants have been developed. Multiple studies have compared the biomechanics of these implants, but there has been little written on clinical outcome differences between them.^70,72

Cages can either be used as a stand-alone device or augmented with other stabilizers. Dorsal instrumentation with pedicle screws through a traditional open approach with or without dorsal fusion is commonly used.^13,15,25,27 Transfacet screws have also been widely used to augment interior interbody fusion.^73,74 In one study translaminar screws were associated with a higher pseudarthrosis rate.⁷⁵ In recent years anterior tension band plates and cages with screw augmentation have also been developed. Although biomechanical studies have compared these implants, there is a lack of comparative clinical studies.^76–80

Motion Preservation Options

In recent years there has been a lot of interest in motion preservation implants using the ventral approach. The best data to date is likely from the FDA trials of the implants that have come to market in the United States. The Charité⁸¹ and ProDisc⁸² trials show short-term results comparable to fusion at 2 years and with limited follow-up at 5 years.⁸³ It is unclear how this technology will fit into our armamentarium in the future.

Author’s Preferred Treatment

In a patient with mechanical back pain and no neurologic symptoms the author’s preferred treatment for this hypothetical 45-year-old male would depend on the underlying process. If the patient had MRI abnormalities only with no disc collapse or instability on radiographs I would likely avoid surgery altogether. I am not sure I understand what the pain generator is in these patients and I am not convinced fusion has a reasonable chance of success.

In the patient with disc collapse and no signs of a spondylolisthesis I would do a stand-alone interbody fusion using a tapered threaded interbody cage with rhBMP-2. This procedure has a high radiographic fusion rate, restores lordosis, has a rapid recovery rate, and avoids any possibility of damage to the dorsal musculature or adjacent facets.

In the patient with an instability I do not feel stand-alone devices are adequate. Although biomechanical studies show ventral plating restores stability, these implants are difficult to place and long-term data is lacking. In this scenario my preference is an anterior interbody fusion with a femoral ring allograft rhBMP-2 and dorsal percutaneous pedicle screw rod fixation. Translaminar facet screws are also a reasonable dorsal option.

Mechanical Low Back Pain without Spondylolisthesis

In this chapter, a 45-year-old male patient with mechanical low back pain is discussed. In the patient with degenerative disc disease and low back pain without spondylolisthesis, fusion surgery is still considered controversial but has been successfully performed when using strict operative indications.^6,7 Anterior lumbar interbody fusion (ALIF) has been reported as an alternative to posterior fusion and has the benefit of decreased blood loss and operative time, as well as avoiding the morbidity associated with the muscle dissection necessary for posterior fusion. However, concerns with the ventral approach include the risk of visceral or potentially life-threatening vascular injury and deep venous thrombosis. In addition, pseudarthrosis rates after ALIF have been reported to be as high as 31%, and retrograde ejaculation can be a significant issue in young males who wish to reproduce.⁸ In addition, graft extrusion requiring high-risk anterior revision surgery is a rare but potentially catastrophic complication of this procedure and of lumbar total disc arthroplasty.⁹ TLIF and PLIF-type procedures are theoretically desirable because they increase the surface area for fusion compared to posterior-alone techniques, but the longer operative times, potential for neurologic injury, and the more recently described potential for postoperative radiculitis and heterotopic bone formation within the spinal canal with the use of bone morphogenetic protein (BMP) are cause for concern.^10,11

Dorsolateral intertransverse fusion, in the appropriately selected patient, can avoid the complications described previously and has been shown to provide acceptable clinical results at mid- and long-term follow-up. Glassman et al. studied 224 patients undergoing single-level dorsolateral arthrodesis with iliac crest autograft for lumbar degenerative disease and found significant improvements in both the Short Form-36 and Oswestry Disability Index scores at up to 2-year follow-up.¹² They observed a fusion rate of 88% to 95% at 2-year follow-up (in younger and older patients, respectively) and reported modest improvements in back pain, leg pain, and return to work. Dimar et al. performed dorsolateral instrumented fusions on 463 patients with lumbar degenerative disease and up to grade I degenerative spondylolisthesis.¹³ They reported fusion rates of 96% with use of rhBMP-2 and 89% with iliac crest and had significant improvements in back pain, leg pain, and return to work status at 2-year follow-up. In conclusion, dorsolateral instrumented fusion can result in significant improvements in pain score and health-related quality of life as well as obtaining a high fusion rate in the appropriately selected patient with mechanical low back pain.

Mechanical Low Back Pain with Spondylolisthesis

In the setting of low back pain secondary to degenerative spondylolisthesis, multiple surgical treatment options have been proposed, including laminectomy without fusion, noninstrumented dorsolateral fusion, and instrumented fusion. In addition, several small case series have been performed looking at the short-term results of interbody techniques, including PLIF and TLIF, but at this time no high-quality randomized, controlled trials exist to advocate these techniques.^14,15

In the 1991 landmark study by Herkowitz and Kurz, 50 patients with lumbar stenosis and degenerative spondylolisthesis were randomly assigned to either decompression alone (25 patients) or decompression with bilateral noninstrumented intertransverse process arthrodesis (25 patients).¹⁶ They reported that 24 patients in the fusion group had excellent or good clinical outcome, whereas 11 patients in the nonfusion group had excellent or good results. In addition, back pain, leg pain, and progression of the spondylolisthesis were all greater in the decompression-alone group.

Although the superior outcomes of dorsolateral fusion versus both nonoperative care and decompression alone have been validated by several studies, the use of instrumentation is more controversial.^17,18 Fischgrund et al. reported on 76 patients randomized to either decompression and fusion without instrumentation or decompression and fusion augmented with pedicle screw instrumentation for single-level degenerative spondylolisthesis with lumbar spinal stenosis.¹⁹ While they noted a significantly greater fusion rate in the instrumented group, this did not translate into improved clinical outcomes, with good or excellent outcomes seen in 76% of the patients with instrumented fusions, and 85% in patients treated with fusion without instrumentation. Similarly, Thomsen et al. reported on 130 patients randomly assigned to either instrumented or noninstrumented posterolateral fusion for Grades 1 and 2 degenerative spondylolisthesis.²⁰ Although they found a slightly higher global satisfaction rate in the instrumented group (not significant), there was significantly higher operative time, blood loss, and early reoperation rate in the instrumented group, while both groups had similar improvement in pain and disability scores. The authors concluded that instrumentation results in similar outcomes to noninstrumented fusion, which does not justify the increased operative times, risk of neurologic injury, and higher reoperation rate. Despite these studies, pedicle screw instrumentation is widely used to increase stability in the early postoperative period during fusion healing, with many arguing that an increased fusion rate results in better clinical outcomes, and therefore if instrumentation increases the fusion rate then clinical outcomes should follow.²¹

Summary

In the setting of mechanical low back pain, both with and without spondylolisthesis, fusion and decompression (if indicated) through a dorsal approach is an acceptable treatment. It has good clinical results while reducing costs and minimizing the risk of complications implicit with more extensive fusion techniques.

Isolated Mechanical Low Back Pain without Spondylolisthesis: Surgical Indications

Chou et al.⁸ recently reviewed 20 randomized trials evaluating surgery for nonradicular low back pain. Of these trials, 4 were prospective, randomized trials evaluating spinal fusion versus nonsurgical therapy for patients with moderate to severe low back pain for at least 1 year and in whom conservative medical management had failed.^9–12 Among these trials were the Swedish Lumbar Spine Study,¹² which showed a statistically significant improvement in patient satisfaction, pain relief, Oswestry Disability Index (ODI) scores, and Visual Analog Scale (VAS) scores, and the Medical Research Council Spine Stabilization Trial,¹¹ which showed a statistically significant improvement in ODI scores. The 2 other studies did not show any statistically significant differences between surgery and nonsurgical therapy at 1 year, although 1 study did show a trend toward better ODI scores and back pain scores in the surgical group.^9,10 Reviews of other similar studies by Gibson et al. showed equally mixed results between surgical and nonsurgical groups.¹³ A possible explanation for this is the lack of standardization of nonsurgical management across the different studies.⁸ Resnick et al al.¹⁴ performed a meta-analysis of several prospective and retrospective studies, examining the performance of lumbar fusion in patients with intractable low back pain without stenosis or spondylolisthesis. They reviewed many of the same studies as Chou et al. and concluded that Class I medical evidence exists in support of the use of lumbar fusion as a treatment standard for the management of isolated low back pain intractable to the best medical management in carefully selected patients.¹⁴ Patients in all 4 studies received a variety of lumbar fusion procedures ranging from dorsolateral fusion alone to interbody fusion supplemented with pedicle screws and posterior noninstrumented fusion (PLF).

Despite the fact that there are only a select few sufficiently powerful prospective randomized trials in the literature, there is significant Class I evidence to support lumbar fusion for isolated low back pain without spondylolisthesis in a certain subset of patients. These patients should have back pain that is intractable to the best medical management, have relatively low comorbidities, and should be well informed of the fact that although most studies show significant evidence that intractable isolated low back pain should improve with fusion, there is still no guarantee of cure.

Isolated Mechanical Low Back Pain with Spondylolisthesis: Surgical Indications

Matsunaga et al.¹⁵ reviewed the outcomes of 145 patients with degenerative spondylolisthesis managed nonsurgically. They found that most of their patients noticed an improvement in their back pain over time, only about 30% of their patients suffered progressive slippage, and 76% of patients who were neurologically intact remained that way. Jacobs et al.¹⁶ reviewed eight prospective, randomized trials and five cohort studies’ management of low-grade adult isthmic spondylolisthesis. They concluded that evidence exists in support of fusion over conservative management for low-grade isthmic spondylolisthesis; however, this should not be considered the gold standard of management. Furthermore, they found no difference in outcomes or complications between circumferential fusion and PLF. Resnick et al.¹⁷ reviewed 85 studies examining fusion and degenerative spondylolisthesis, among them the landmark study by Herkowitz and Kurz,¹⁸ which provided Class II evidence in support of decompression in conjunction with fusion in patients with lumbar stenosis with spondylolisthesis versus decompression alone. They concluded that lumbar decompression with fusion is effective in the management of spinal stenosis associated with spondylolisthesis. Sengupta and Herkowitz¹⁹ listed three indications for surgical management of degenerative lumbar spondylolisthesis: persistent back pain and/or leg pain refractory to 3 months of the best medical management and significant reduction of quality of life (QOL), progressive neurologic deficit, and bladder or bowel symptoms. The presence of spinal stenosis with spondylolisthesis is important; several trials (including the much-discussed Spine Patient Outcomes Research Trial [SPORT]) indicate that surgical management is clearly helpful if medical management fails.²⁰

There is much controversy in the literature over optimal management of adult degenerative and isthmic spondylolisthesis associated with isolated back pain. As with isolated back pain alone, however, a solid body of evidence supports the use of spinal fusion in the management of isolated low back pain with spondylolisthesis in patients in whom a trial of best medical management has failed and who have persistent low back pain and reduction of QOL.

Current Surgical Options

The combined ventral and dorsal approach comprises two procedures: ventral fusion by means of discectomy and interbody graft placement and dorsal instrumentation with or without decompression and/or fusion. Ventral fusion can be achieved in a variety of ways, the most common being anterior lumbar interbody fusion (ALIF), posterior lumbar interbody fusion (PLIF), transforaminal lumbar interbody fusion (TLIF), or the lateral transpsoas approach to interbody fusion (LTIF). These procedures provide solid anterior column support via a bone graft or cage (commonly titanium mesh or polyetheretherketone [PEEK] cages filled with osteoinductive filler). These procedures can then be combined with a dorsal decompressive laminectomy, PLF, facet osteotomy, and/or pedicle screw placement. Combining ventral fusion procedures with a PLF is commonly referred to as circumferential or 360-degree fusion, whereas combining ventral fusion with posterior column support alone (e.g., via pedicle screws) is called 270-degree fusion.

A combined dorsal and ventral approach has several advantages over a single ventral or dorsal approach: it provides the best chance for bony fusion and the lowest rates of pseudarthrosis, and it allows the surgeon the most degrees of freedom to restore sagittal balance, reduce listhesis, and decompress the neural elements while distracting the interspace. However, due to its considerable associated complication rate, it should be reserved for the carefully selected, appropriate patient.

Combined Ventral and Dorsal Approach

Strong evidence exists in support of spinal fusion as an option for the management of adult isolated low back pain with or without spondylolisthesis refractory to best medical management. Many of the major randomized, controlled trials did not specifically evaluate the use of one technique over another, and most of the studies in the literature lack sufficient power to strongly suggest one technique over another.

Mechanical low back pain is thought to be, in part, a result of intervertebral disc degeneration and dehydration of the nucleus pulposus, which results in altered distribution of load across the vertebral endplates.²¹ This explains why pain is worsened with daily activity (loading) and improved with rest (unloading). Although it is commonly thought that spinal instability in mechanical back pain and degenerative spondylolisthesis is a result of abnormal translation and motion at the affected segment, there is minimal evidence to support this.¹⁹ As a matter of fact, it has been shown that motion at the diseased segment is actually decreased in patients with degenerative spondylolisthesis.²² However, proponents of the combined approach hold that removal of the pain-generating disc and solid support of the anterior column with an interbody graft combined with posterior column fusion (with or without pedicle screws) provides the best chance for spinal fusion and optimal outcomes when compared to PLF alone.

Resnick et al.^23–25 performed meta-analyses to study the use of fusion as an adjunct to decompression in patients with low back pain, with and without spondylolisthesis. Based on the medical evidence they analyzed, they concluded that PLF and interbody fusion (via TLIF, PLIF, or ALIF) should be considered as treatment options in patients with low back pain and degenerative disc disease at one or two levels. Interbody techniques are associated with higher fusion rates when compared to PLF alone, although the medical evidence supporting this statement is marginal. Furthermore, they state that Class II evidence exists to suggest that outcomes are improved in those patients who receive an interbody graft, and no Class I or II evidence exists suggesting worse outcomes with interbody grafts. They do stress that complication rates are significantly higher with the combined approach, and there is no evidence to support the position that outcomes are improved with the addition of a PLF to an ALIF or ALIF with pedicle screws, as this will simply add to morbidity and cost.

Kwon et al.²⁶ performed a systematic literature review to examine outcomes in management of adult low-grade isthmic spondylolisthesis. They examined 34 studies, 4 of which were randomized, controlled studies. They found that patients undergoing a combined approach attained significantly higher fusion rates (98.2%) when compared to ventral (74%) or dorsal (83.3%) procedures alone, and had a higher rate of successful clinical outcome (86.4%) versus dorsal (74.8%) or ventral (89.6%) procedures. In many of the studies that they reviewed, there seemed to be a strong correlation between achievement of a solid fusion and successful clinical outcome. The authors do state that their data was based on pooled results from many different studies (many of which were retrospective) and therefore may be subject to a certain level of bias. Especially alarming is the fact that two of the better randomized studies showed no difference between the two groups.

Swan et al.²⁷ performed a prospective randomized trial comparing single-level PLF to the combined approach in patients with unstable isthmic spondylolisthesis. They found superior clinical and fusion outcomes in patients undergoing a circumferential procedure compared to the PLF group. There was also a higher rate of complications in the circumferential group.

Videbaek et al.²⁸ performed a prospective, randomized study of circumferential fusion versus instrumented PLF in patients with severe chronic low back pain. They found significantly better clinical outcomes in the circumferential group when compared to the instrumented PLF group at 5 to 9 years after surgery.

Conclusions

Mechanical low back pain is a common affliction in the Western world. Mechanical low back pain may be associated with isthmic or degenerative spondylolisthesis. In patients in whom a course of conservative medical management has failed, lumbar fusion of affected motion segments has been shown to be a viable therapeutic option. A combined dorsal and ventral approach to lumbar fusion is associated with high radiographic fusion rates, low incidence of pseudarthrosis, and excellent long-term clinical outcomes. There is no evidence to suggest that this procedure is more optimal than or inferior to ventral or dorsal fusion alone in terms of clinical outcomes.

Several different methods by which a combined approach may be accomplished are available. The instrumented ALIF with dorsal fusion is very successful in terms of fusion rates and patient satisfaction, but it requires two separate procedures and may be precluded depending on patient pelvic anatomy. PLIF is a traditional method of dorsal circumferential approach; TLIF is a newer, less invasive method. Both procedures show similar successful outcomes, but many surgeons favor the shorter, midline-sparing TLIF. LTIF is the newest procedure for circumferential combined approach. It too requires two separate procedures (although some surgeons have seen success in the short term with stand-alone LTIF) and is associated with unique, undesirable complications, such as lumbar plexus injury. It does show promise as a minimally invasive approach to interbody fusion.

There is great debate as to whether or not circumferential fusion is indicated for back pain with or without spondylolisthesis. Most studies show that it adds potential morbidity and complications and has the same fusion rates as instrumented ventral fusion alone.

Discectomy

Discectomy is a well established method for treating radicular pain; however, its success ranges between 48% and 89% in the literature.² Long-term follow-up of patients who underwent lumbar discectomy reveals that 50% to 60% will suffer significant back pain after 10 years and 20% to 30% will develop recurrent radicular pain.² These outcomes are clearly suboptimal and likely due to a combination of ongoing degeneration, segmental instability, spinal stenosis, and recurrent disc herniation.

Reasons for long-term failure of surgical treatment include failure to address the primary pain generator and collapse of the disc space. Disc space collapse can lead to a decrease in the diameter of neural foramina, alterations in facet loading, and disruption of sagittal balance. The abnormal loading of the joints results in biochemical changes in the cartilage that can affect both the diseased level and adjacent segments. Chronic abnormal loading can also trigger hypertrophic changes of the articular processes. The disc space narrowing has the potential to cause rostral and anterior displacement of the superior facet joint, which may compress the exiting nerve root as it traverses the foramen.²

Ongoing pain may also result from destabilization of the spinal segment after discectomy. Motion segment stiffness is inversely related to the amount of nuclear material that is removed, and removal of even a portion of the disc has been associated with hypermobility in the sagittal, coronal, and axial planes.^2,3 This instability can eventually extend to the rostral adjacent level and exacerbate the symptoms.^2,4

Arthrodesis

Arthrodesis eliminates segmental instability, preserves sagittal balance, stops further degeneration at the treated level, and is able to restore disc space height with an interbody graft. For these reasons, it has become the primary treatment for symptomatic spondylosis. Positive results have been reported in 52% to 100% of patients who underwent anterior lumbar interbody fusion (ALIF) and in 50% to 95% of those who underwent posterior lumbar interbody fusion (PLIF).^2,5,6 Despite these results, spinal arthrodesis has potential negative consequences. Fusing a spinal segment disrupts the biomechanical mobility of adjacent levels and potentially the entire spinal region. Over time this may lead to symptomatic hypertrophic facet arthropathy, spinal stenosis, disc degeneration, and osteophyte formation.^2,7 Long-term follow-up of lumbar fusion patients has demonstrated that approximately 50% eventually develop postoperative low back pain, and approximately 15% may require additional surgery.^2,7 In the cervical spine, hypermobility at segments adjacent to an anterior fusion construct has been repeatedly demonstrated and likely accelerates the degenerative process. Long-term follow-up studies show radiographic adjacent-level disease in up to 92% of patients after anterior cervical fusion.⁸ Although these findings do not correlate well with clinical symptoms, the accelerated spondylotic changes are becoming an increasing concern with this procedure.

Other problems associated with fusions include the potential for fixation with suboptimal sagittal balance, morbidities associated with autograft harvest, and disruption of muscular and ligamentous attachments. Although the loss of mobility is also a drawback, this has a greater functional impact in the cervical than in the lumbar spine.⁹

Arthroplasty

The concept of reconstructing a functional motion segment after decompressing impinged neural elements is not new. The first attempts at disc arthroplasty began more than 40 years ago.^2,10 Ulf Fernstrom made the first attempt to reconstruct the disc space by using stainless steel balls in both the lumbar and cervical spine. He reported his results in 1966 but unfortunately this reconstruction method had a high rate (88%) of subsidence.¹⁰ Other early attempts were also associated with unacceptable complication rates involving migration and subsidence of the device as well as adjacent-segment hypermobility.¹¹ As advancements have been made in our understanding of the degenerative process, the biomechanics of motion, and the biomaterials available for design, the possibility of replacing a degenerated disc with a functional, stable joint has become a reality.

Advantages and Indications

Disc arthroplasty is advantageous over arthrodesis because of its ability to restore spinal segment mobility by mimicking the kinematics of the intervertebral disc while maintaining stability and normal physiologic loads (Fig. 224-3).

FIGURE 224-3 Lateral radiographs demonstrating preserved mobility in flexion and extension after implantation of the cervical ProDisc.

(Courtesy of Synthes Spine.)

In terms of neurologic outcome, pain resolution, and functional mobility, arthroplasty offers equivalent if not superior results to segmental fusion in the cervical spine. Other advantages include the elimination of the risk for developing pseudarthrosis. The risk of pseudarthrosis after a fusion procedure is estimated to be as high as 10% and can result in chronic pain and pathologic loading of surrounding joints.^5,6 Many advantages that have been associated with fusion techniques, such as elimination of future disc herniation, are also expected with arthroplasty.¹⁶ Expected positive outcomes similarly include a reduction in development of spondylosis, stenosis, and instability at the treated segment. The restoration of intervertebral height will increase foraminal diameter and diminish the occurrence of nerve root compression, and the morbidities associated with autograft harvest are avoided entirely.

Indications for undergoing disc arthroplasty in the cervical spine are similar to those for fusion. Any patient in whom conservative treatment for radicular pain has failed and who has radiographic evidence of disc disease is a candidate. Patients with symptomatic spondylosis are also candidates. Low back or radicular pain in the setting of radiographic evidence of single-level disc disease is an indication for lumbar arthroplasty.

Disadvantages

Not all patients with spinal disc disease are candidates for arthroplasty. Contraindications are numerous and include segmental instability, prior trauma resulting in compromise of bony structure, severe spondylosis with absence of motion and bridging osteophytes, infection, and presence of other spinal diseases, such as osteoporosis, rheumatoid arthritis, and known malignancy involving the spine. The presence of significant deformity such as scoliosis, spondylolisthesis, or kyphosis, is also a contraindication.^16,17 Patients who have preoperative hypermobility that exceeds 11 degrees of flexion and extension are likely poor candidates for arthroplasty, because long-term stability may not be successfully maintained and subluxation or dislocation of the disc replacement can occur.¹ A known allergy to any of the materials found in the implant eliminates this procedure as a surgical option, and a potential reaction to one of these elements is another disadvantage of the procedure.

Cadaveric studies assessing the facet force pressures before and after both cervical arthroplasty and cervical fusion found that although the changes in facet force were minimal in flexion, lateral bending, and rotation, they were significantly increased in extension at the treated level in the arthroplasty group.¹⁸ Findings such as these have precluded patients with facet joint disease from being acceptable candidates for disc arthroplasty and have raised concerns about escalation of degenerative changes in the normal facet joints in those patients who have undergone the procedure.

Fundamental Concepts

Nucleus Replacement

Some of the initial efforts at disc restoration focused on replacing the nucleus pulposus. Nucleus replacement is a less invasive option than fusion or total disc replacement for treating degenerative disorders. These devices avoid major disruption of the anulus fibrosus because they can be placed through a small anular window. Advantages to nucleus replacement include multiple approach options, multiple revision options, and minimally invasive technique. Disadvantages include the risk for migration, expulsion, or subsidence. Contraindications to this type of procedure include an intervertebral space less than 5 mm, spondylolisthesis, and presence of Schmorl nodules. These contraindications unfortunately limit the usefulness of this technique as a restorative procedure for spinal degeneration.²⁵

Indications for nucleus replacement can be categorized as prophylactic after microdiscectomy or as a treatment for low back pain in the setting of early to moderate degenerative disease with significant degeneration of the nucleus but an intact anulus. If the nucleus is the presumed pain generator, its removal and subsequent reconstruction with a replacement may improve symptoms and maintain functional motion. Alternatively, loss of disc tissue from herniation or microdiscectomy and its associated intervertebral height loss can accelerate degeneration of the spinal segment. Prophylactic placement of a nucleus prosthetic could restore disc height, absorb physiologic load, and deform with movement while maintaining structural integrity.²⁵

A variety of nucleus replacements are being developed and studied. These devices can be classified as either elastomeric or mechanical. The elastomeric devices frequently consist of hydrogel polymers that absorb water and swell when placed in the intervertebral disc. Nonhydrogel polymers are low-friction materials that carry shock absorption characteristics. This category of implant is theoretically able to recreate the natural function of the normal nucleus pulposus with uniform stress distribution and the ability to absorb shock. Some of these devices are preformed and some are injectable. Preformed devices are deformable and therefore more likely to extrude; however, injectable devices can leak and carry concerns regarding their durability and biocompatibility. Mechanical nucleus replacements are designed with either one or two pieces and are constructed from various biomaterials, including metal, pyrolytic carbon, and PEEK.²⁵ This type of device has more strength and durability than the elastomeric devices; however, it is not able to maintain an even stress distribution and lacks shock absorption capacity. A predisposition to subsidence and expulsion is also problematic given the lack of a mechanical anchor.^25,26

The most widely used nuclear replacement is the Prosthetic Disc Nucleus (PDN; Raymedica, Minneapolis, MN). This elastomeric device consists of a hydrogel core constrained within a woven polyethylene jacket. One or two devices are placed in the disc space in an attempt to restore disc space height and replicate physiologic function. The jacket is flexible but inelastic and functions to constrain horizontal and vertical expansion and deformation of the hydrogel core. The core is designed to expand on hydration after implantation. This expansion occurs over 4 to 5 days, during which time the nucleus prosthesis widens the interspace.^2,27,28 Biomechanical studies have demonstrated that this device is able to reproduce the range of motion and stiffness of a healthy spinal segment. The neutral zone is also maintained within physiologic limitations, providing stability.^28,29 Short-term follow-up studies have demonstrated the effectiveness of this device, with successful maintenance of disc height and symptomatic relief seen in 88% of patients. Long-term follow-up, however, has not been as encouraging. The results are inferior to those of arthrodesis, and the complications rates are high.^27,29 The tendency for device migration is especially concerning, and iatrogenic injury to the surrounding ligaments that may constrain the device needs to be minimized. Other preformed elastomeric devices under evaluation include the NeuDisc (Replication Medical, Cranbury, NJ) and the Percutaneous Nucleus Replacement (PNR; TranS1, Wilmington, NC). Injectable devices under evaluation include BioDisc (CryoLife, Atlanta, GA), Hydrafil (Synthes, West Chester, PA), and NuCore (Spine Wave, Shelton, CT).

Examples of mechanical nucleus replacements include the Nubac Disc Arthroplasty Device (Pioneer Surgical, Houten, The Netherlands), a two-piece mechanical nucleus constructed with PEEK, and the Regain device (Biomet, Warsaw, IN), a one-piece nucleus constructed of graphite and coated with pyrolytic carbon. Both are being evaluated for clinical use.

Placement of a nucleus replacement device should only be undertaken in the setting of a valid indication and after weighing all other treatment options. Limited clinical information is available for these implants and they remain investigational despite some early encouraging results.^26,30 Further clinical investigation is needed to determine if this technology has the capacity to slow down degenerative changes in the spine.

Total Disc Arthroplasty

Total disc arthroplasty is a growing strategy for treating cervical and lumbar radicular and discogenic pain resulting from degenerative disease. An optimal disc replacement will stabilize and preserve motion of the functional spinal unit after successful decompression of the neurologic structures and avoid long-term acceleration of degenerative changes at adjacent segments.

Four artificial discs are currently approved by the Food and Drug Administration (FDA) for single level use in the United States. These include the lumbar Charité (DePuy Spine, Raynham, MA), the cervical and lumbar ProDisc (Synthes, West Chester, PA), the cervical Prestige (Medtronic, Fridley, MN), and the Bryan Cervical Disc (Medtronic). The Kineflex-C and Kineflex-L devices (SpinalMotion, Mountain View, CA), PCM cervical disc (NuVasive, San Diego, CA), and the CerviCore disc (Stryker Spine, Allendale, NJ) are currently being evaluated for FDA approval. Numerous other devices are in various stages of development or clinical use outside of the United States (Table 224-1).

Artificial Disc Replacement Devices

Charité

The Charité device is the artificial disc with the longest history of implantation and clinical investigation.² Developed by Karin Buttner-Janz and Kurt Schellnack at the Charité University Hospital in Berlin in the mid-1980s, it was approved for use in the United States in 2004. The disc provides unconstrained kinematic motion of the spinal segment and consists of a three-piece articulating device with a biconvex polyethylene core spacer interfacing with two CoCrMo alloy endplates. The device endplates have dorsal and ventral teeth that anchor them into the bony endplates and prevent migration of the implant. Titanium and hydroxyapatite coat these endplates to promote bonding with the vertebral bodies.^2,31,32

Five-year follow-up data from the FDA Investigational Device Exemption (IDE) prospective, randomized trial has been published for the Charité artificial disc. Comparison with ALIF was made at a single level from L4-S1. Overall success was 57.8% in the Charité group and 51.2% in the ALIF group. No major complications or neurologic deteriorations were demonstrated. Given the lack of statistical differences in clinical outcome, the device has been deemed a safe and effective alternative to fusion. Encouraging secondary outcomes included greater rates of part- and full-time employment and lower rates of long-term disability among patients who underwent placement of the Charité disc when compared to those who underwent an ALIF procedure.³²

Because the Charité was the first implant to be used for total disc replacement, the longest follow-up data is available with this device. Retrospective data over an average of 17 years reveal a 60% rate of ankylosis at the treated level. Those patients who retained mobility had no observable evidence of adjacent-level disease; however, this group was significantly less satisfied with their outcome than those patients who had undergone fusion, and 11% required reoperation. These long-term results did not provide any clear evidence that total disc replacement in the lumbar spine is a superior treatment to spinal arthrodesis.³³

ProDisc

Dr. Thierry Marnay developed the ProDisc system in the late 1980s in France. The ProDisc has been designed for both the lumbar spine (ProDisc-L) and the cervical spine (ProDisc-C). This artificial disc is constructed based on a ball-and-socket principle. A three-piece articulating device, it consists of a monoconvex polyethylene inlay and two CrCoMo alloy endplates. The inlay locks into the inferior endplate, and both endplates have a central keel that secures them to the vertebral bodies to prevent migration. Titanium Plasmapore coating of the endplates promotes bony integration. The monoconvex nature of the polyethylene inlay allows the implant to be inserted with less distraction than the Charité device. This inlay is inserted into the caudal endplate, allowing only two moving parts after implantation. This results in semiconstrained kinematic behavior (Fig. 224-4).²

FIGURE 224-4 ProDisc device demonstrating mobility of the endplate over the monoconvex inlay during flexion and extension.

(Courtesy of Synthes Spine.)

The safety and efficacy of the ProDisc-C for treatment of symptomatic cervical disc disease has been established in a prospective, randomized study that compared use of the device to ACDF. Symptomatic outcome is not significantly different between the two procedures; however, after 2-year follow-up the rate of reoperation was significantly lower in the ProDisc-C group (1.8% compared to 8.5%). The use of narcotic pain medication was significantly lower in the disc replacement group as well, and 84.4% of patients with artificial discs retained at least 4 degrees of motion in flexion-extension at the treated segment after 2 years.^34,35

A similar study with the ProDisc-L device in the lumbar spine demonstrated safety and efficacy in comparison to circumferential fusion. Lumbar disc arthroplasty was found to have significantly superior results when considering the Visual Analog Scale (VAS) for pain assessment. After a 2-year follow-up, 93.7% of patients retained a normal functional range of motion, with an average of 7.7 degrees mobility on flexion-extension films.^36,37

Bryan

The Bryan Cervical Disc was created in the 1990s by the American neurosurgeon Vincent Bryan. This disc is designed to allow for normal, unconstrained physiologic cervical motion. The device consists of a polyurethane nucleus that is shaped to fit between two dome-shaped titanium shells. The shells have a rough-textured coating to encourage osseous ingrowth. A flexible polyurethane membrane surrounds the nucleus and prevents debris migration or ingrowth of connective tissue. A tight fit into a milled concavity is required to prevent migration because no hardware anchors the implant. This device allows for normal cervical range of motion in flexion, extension, lateral bending, axial rotation, and translation as well as providing shock absorption of axial loads (Fig. 224-5).^2,38

FIGURE 224-5 Bryan Cervical Disc.

(Courtesy of Medtronic.)

Prospective, randomized studies have demonstrated that the Bryan disc is safe, effective, and compares favorably to cervical fusion for treatment of single-level disease. Patients who underwent arthroplasty showed significant improvement with the Neck Disability Index and VAS for neck and arm pain and returned to work almost 2 weeks earlier than those who underwent fusion. An average of 7.9 degrees of flexion-extension motion was retained after 2 years.^39,40

Prestige

B.H. Cummins created the initial version of this artificial disc at the Frenchay Hospital in Bristol, UK, in 1989. After initial disappointing outcomes, the disc was redesigned and reintroduced after pilot studies demonstrated favorable results. In its modern version, the Prestige cervical disc provides unconstrained motion at the treated spinal segment.¹¹ The initial device was constructed with stainless steel; however, the latest generation is a titanium and ceramic composite that is MRI compatible. A porous titanium plasma spray coating is present to encourage bone ingrowth. The disc consists of two articulating components in a ball-and-trough configuration. A ventral extension of each component anchors into the vertebral body with screws. This implant aims to maintain physiologic degrees of flexion, extension, lateral bending, and rotation (Fig. 224-6).

FIGURE 224-6 Prestige disc device.

(Courtesy of Medtronic.)

The prospective, randomized trial for the Prestige cervical disc demonstrated safety, efficacy, and preservation of segmental sagittal motion after 2 years with an average of more than 7 degrees of movement. Clinical outcomes were improved when compared to ACDF, and lower rates of secondary surgeries were seen in the arthroplasty group (Fig. 224-7).⁴¹

FIGURE 224-7 Lateral radiograph demonstrating the Prestige cervical disc device.

(Courtesy of Medtronic.)

Complications

Potential complications will always coincide with the development of new operative techniques. Many of the potential adverse events that can occur with cervical or lumbar fusion are also risk factors for arthroplasty. These include CSF leak, postoperative hematoma, dysphagia and hoarseness (cervical spine), and injury to visceral or neurovascular structures. Complications more specific to total disc replacement include intraoperative or delayed migration of the implant, subsidence, device failure leading to segmental hypermobility or increased stiffness, heterotopic ossification, accelerated facet degeneration, and spontaneous fusion. Thus far, major operative and device-related complications have been infrequent. In the lumbar spine, arthroplasty has been associated with a device-related complication rate that ranges from 0 to 6.5%. The total perioperative complication rate after cervical arthroplasty has been reported at 6.2% per treated level.¹

Adverse outcomes are frequently associated with inappropriate sizing or poor placement of implants. Overdistraction of the facet joints can lead to postoperative neck pain; careful preoperative assessment of disc space height should guide the surgical plan. Unfortunately, no clear data are available to clarify how much height loss would invalidate arthroplasty as a treatment option. A negative effect on sagittal alignment in the cervical spine has also been observed in scattered cases with reports of postoperative kyphotic deformity. Although this effect could have significant implications for long-term outcomes and rates of adjacent-level disease, no precise avoidance method has been determined. Factors that might affect the sagittal alignment include the angle of disc insertion, bone removal, and endplate preparation.¹

Should disc arthroplasty fail, arthrodesis at the affected level is the appropriate treatment. Described revision strategies include fusion with posterior instrumentation while leaving the disc device in place or anterior explantation with conversion to arthrodesis. Success of fusion after removal of the disc prosthesis is slightly better than with posterior fusion alone.⁴² Appropriate patient selection will help to minimize complications.

Conclusions

The definitive objective of a disc arthroplasty device is to stabilize the treated segment after successful decompression of the neurologic structures, preserve physiologic motion, and mimic the kinematics of a normal, healthy intervertebral disc. Biomechanical studies continue to demonstrate that disc replacement creates less adjacent-level strain and hypermobility than fusion. Results from clinical studies have thus far been encouraging, and evidence is mounting that cervical disc arthroplasty is a superior treatment strategy to ACDF. Although the outcomes after lumbar arthroplasty have not been as encouraging as those after cervical arthroplasty, further evolution of these implants and their potential to interact with dorsal dynamic devices could enhance their appeal as a surgical treatment option in the lumbar spine.

As our understanding of spinal kinematics and degenerative disease improves and our testing methodologies are refined, the potential for disc arthroplasty will expand, enhancing our ability to treat patients with spinal disease.

Clinical Management

When considering surgical treatment with a NR device, the same standard clinical investigations should be performed as for fusion or total disc replacement (TDR). The well-known exclusion criteria for TDR (e.g., osteoporosis, infection) also apply for NR.

Logically, the goal of NR is to eliminate pain and restore or maintain the disc height and natural load sharing between the nucleus and anulus to achieve mechanical stability. This requires some different considerations than TDR or fusion, because a NR device biomechanically relies much more on cooperation with the biological structures adjacent to the implant.

Standard Anteroposterior and Lateral Radiographs

Imaging in standing position, flexion, and extension (only for motion preservation) could help with basic filtering of treatment options, because the presence of spondylolisthesis and disc height could be evaluated. The critical requirement for a NR procedure versus interbody fusion or TDR is a certain minimum disc height. Disc height loss typically indicates late-stage disc degeneration; therefore, disc height could be considered an indication for the stage of the disc degeneration cascade. Disc height loss is also combined with increased shear forces to the anulus fibers and thus their loss of mechanical stability.

Also, most NR devices require a minimum average disc height of 5 mm to insert the implant and avoid overstretching the anulus. This overstretching can reduce the ability to regain normal disc height once the implant is in place. It also might lead to increased contact stress between the device and the endplates with the risk of subsidence or anulus rupture with dislocation. Therefore, late-stage degeneration with disc height less than 5 mm is considered an indication for TDR or fusion rather than NR.

Magnetic Resonance Imaging of the Lumbar Spine

MRI has nearly become routine in recent years, providing important insight into the degenerative state of the discs and the spine in general. Important classifications of the disc and endplates are the Pfirrmann and Modic classifications, which help determine the degenerative state of the target disc and evaluate the adjacent discs in consideration of potential multilevel treatment.

Fluoroscopically Guided Diagnostic Injections

To further discuss the available options in this case without spondylolisthesis, an ongoing and very passionate clinical controversy needs to be re-addressed; namely, whether test methods beyond conclusive radiographic and MRI should be utilized in the diagnosis of mechanical low back pain, especially if they are as invasive as provocation discography. With the disc and/or facet joints as the usual suspects for the source of pain, not only two different sources, but also two different treatment pathways need to be considered. The predominant current clinical opinion excludes patients with significant facet pain based on morphologically altered arthritic joints from treatment with a TDR or NR arthroplasty. For a clinical workup in anticipation of a surgical treatment, fluoroscopically guided injections for the facets (medial branch blocks) and potentially a provocation discography seem helpful. Aside from the pain response (provocation part) during the discography procedure itself, it is possible with subsequent CT of the disc to rate the degree of internal anulus disruption. The degree of disruption is the underlying cause for discogenic pain according to the current concept¹ and is a sign of mechanical instability. Hence, it should not exceed a certain value to guarantee a functional anulus, one with the ability for load sharing with a NR device and also able to prevent its dislocation (data on file, unpublished).

Thus, a diagnostic workup that includes fluoroscopically guided injections might not always be absolutely necessary but often provides very helpful additional information.

Available Nucleus Replacement Devices

The use of NR devices in patients with mechanical back pain in the earlier stages of degeneration and with only moderately disrupted anuli has been investigated in several recent studies,² including results for an injectable device (DASCOR System, Disc Dynamics, Eden Prairie, MN)³ and a mechanical device (NUBAC System, Pioneer Surgical Technology, Marquette, MN).⁴ Both have demonstrated their effectiveness and safety profiles. As of the end of 2009, several NR designs were in preliminary clinical investigation in the United States; however, only the NUBAC System is currently in an active Investigational Device Exemption (IDE) Food and Drug Administration (FDA) pivotal trial. Other NR designs are still under development or have been discontinued for technical reasons and/or adverse events. In Europe, certain NR devices have gained CE Mark approval and are still available. The different nucleus designs can be classified according to their material, their biomechanical behavior, or their surgical insertion features (Fig. 224-8). Chapter 163 has a more comprehensive overview of all NR devices.

FIGURE 224-8 Classification of nuclear replacement devices.

(Data provided by Coric D, Mummaneni PV: Nucleus replacement technologies. J Neurosurg Spine 8:115–120, 2008.)

DASCOR Disc Arthroplasty System

The DASCOR device is a two-part in situ–cured polyurethane injected within an expandable polyurethane balloon to form the final NR device, which conforms to the nucleus cavity created (Fig. 224-9). The pressurized injection of the device into the disc space is controlled using a proprietary injection system.

FIGURE 224-9 Spine model of the DASCOR Disc Arthroplasty System. An in situ-cured polymer is injected into the disc space within a containment balloon. The expandable balloon fills the space that has been created during the complete nuclectomy.

(Courtesy of Michael Ahrens, MD.)

Before the implantation, the system has the ability to verify intraoperatively the volume and size of the resultant DASCOR device using fluoroscopic imaging procedures in conjunction with a proprietary DASCOR imaging balloon. This imaging system also gathers important information about anulus integrity, enabling it to be used as a stand-alone system in conjunction with a discectomy procedure (Fig. 224-10). For the long-term functionality of a NR device, information about anulus integrity is vital. If the imaging step reveals an incompetent anulus, an alternative procedure for stabilization could be chosen. It is believed that many of the inherent design features of the DASCOR system contributed directly to its positive initial clinical results. For example, the combination of a pliable modulus of elasticity and a large cross-sectional area and surfaces that contour to the endplates supports the goal of creating a uniform load distribution on the endplates. Also, by filling the entire surgically created nucleus space under controlled pressure, load sharing between the nucleus and the anulus (prosthesis and endplates) is re-established. Migration is unlikely because only a small anulotomy is required for implantation of the catheter balloon; the device achieves its full size only after in situ injection. If ever required, revision can be easily performed.

FIGURE 224-10 Anulus assessment with the DASCOR imaging step. To verify and measure the prepared nucleus cavity for an injectable device like the DASCOR, a balloon is filled with contrast medium and visualized under fluoroscopy. Here the anteroposterior and lateral images show the disc space symmetrically filled with the contrast fluid, indicating an appropriate removal of the disc material.

(Courtesy of Michael Ahrens, MD.)

The DASCOR device gained CE Mark approval in Europe in 2005 and an enhanced CE Mark for minimally invasive use via a dorsolateral tube approach and endoscopic access in 2009. Its manufacturer (Disc Dynamics, Inc.) finished enrollment in its U.S. IDE pilot feasibility study, but business operations ceased in early 2010 because of the prolonged FDA approval process to initiate a pivotal study and resulting financial difficulties. Continuation of the project and use of the technology is unclear.

NUBAC Disc Arthroplasty System

The NUBAC device is a two-piece mechanical nucleus. It has a ball-and-socket design with a large surface contact area to achieve load sharing, enable uniform stress distribution, and theoretically decrease the risk of subsidence and extrusion. The system is composed of PEEK-OPTIMA material, which offers well-established biocompatibility, biodurability, wear resistance, and radiolucency. The NUBAC is the first PEEK-on-PEEK articulated intradiscal arthroplasty device (Fig. 224-11).

FIGURE 224-11 NUBAC, artificial two-part nucleus replacement device. Both parts of the device are composed of PEEK with a ball-and-socket design with a large surface contact area to distribute stress. The insertion is possible through several minimally invasive surgical approaches.

(From Pioneer Surgical Technology, Inc. Used with permission.)

More than 300 devices have been implanted worldwide since 2004, including 20 patients as part of an FDA IDE pilot feasibility study. In 2008, the FDA granted permission to start a full pivotal study in early 2009 with 400 patients randomized 1:1 against total disc arthroplasty (ProDisc-L, Synthes, West Chester, PA). The indication is symptomatic single-level degenerative disc disease in the L4-5 level only, defined as discogenic back pain with or without leg pain, with degeneration of the disc as confirmed by patient history, physical examination, and radiographic studies showing decreased disc height, contained herniated nucleus, vacuum phenomenon (dark disc), or positive discography at the affected level.

NUBAC has CE Mark in Europe and has been implanted through a lateral transpsoas approach at L3-4 and L4-5 as well as through ventral retroperitoneal and dorsal approaches at L5-S1.

Conclusion

Spondylolisthesis is currently considered an exclusion criterion for nucleus replacement. Therefore, this 45-year-old with mechanical back pain, without spondylolisthesis and no neurologic findings, needs to have a moderately intact anulus for effective treatment with a nucleus replacement. Primary surgical access could be lateral or dorsal via a minimally invasive, tube or endoscopic system, leaving revision strategies open. Without a preoperative or intraoperative assessment of the anulus, the use of all NR designs in patients subsequent to a discectomy has either failed or remains clinically unproven. The design of the NUBAC device may have this concern solved given its mechanical properties. However, for the elastomeric NR devices, either a preoperative assessment with postdiscography CT or an intraoperative anulus assessment similar to the DASCOR imaging technique could overcome this limitation. Whether assessment of the anulus needs to be made preoperatively and whether or not discography is the right tool still needs to be proven in larger clinical studies. At present, conclusive imaging and an intraoperative anulus assessment with a valid treatment alternative seem reasonable.