25

The prevalence of lumbar disk degeneration is estimated to be present in 15 to 45% of individuals in the United States.1 Although the vast majority of individuals affected by radiographic evidence of degeneration have none or minimal symptomatology, there is a population of patients that suffer from debilitating mechanical low back pain. Pain stemming from lumbar degenerative disk disease (DDD) is multifactorial, and it is frequently difficult to identify the exact source of a pain generator with current diagnostic applications. Controversy arises from the fact that surgery is often performed on patients who have degenerative changes in the lumbar spine, and their clinical outcomes are variable.

Chronic low back pain (CLBP) is a significant cause of disability of younger and middle-aged individuals and has a considerable socioeconomic impact based on cost of treatments and duration of lost wages from disability, among other issues.² The consensus for treating patients with a rational approach is still lacking, although nonoperative approaches remain the mainstay among all options. While nonoperative treatment options are not standardized, it is still accepted that in the absence of progressive neurological injury or compromise, infection, malignancy, or significant spinal instability, this course of action should be initially adopted. At least 6 to 12 months of conservative care should be offered to the patient with CLBP, including combinations of exercise, physical and/or manual therapy, medications, and behavioral and activity modifications. Although none of these options is proven to be the most effective approach, they are often used in combinations to restore health and function and to limit the amount of pain and disability that a patient with CLBP is experiencing.

Failure to improve with a conservative treatment program for a nonspecified amount of time often results in surgical consultation. As was stated previously, surgery for CLBP is controversial due to the variable clinical results and indications; however, the standard of care in 2009 for the management of CLPB from lumbar disk degeneration remains arthrodesis. Spinal fusion is based on the concept that there is potentially either abnormal motion or bio-mechanical load in the degenerative disk segments leading to pain symptoms, and elimination of these problems will perhaps improve pain. Modalities including magnetic resonance imaging (MRI) or computed tomography (CT), plain radiographs, provocative diskography, diagnostic facet blocks, and psychosocial evaluation are all incorporated into the diagnostic schema to identify potential pain generators and to recognize other sources of pathology leading to the symptoms.

Numerous techniques for spinal fusion exist and are continuously being modified and adapted to the specific anatomical problems being addressed (i.e., number of degenerative levels, specific levels involved, presence or absence of instability, etc.). Surgical options include posterolateral fusion with or without instrumentation, posterior lumbar interbody fusion (PLIF), transforaminal interbody fusion (TLIF), anterior lumbar interbody fusion (ALIF), and circumferential, 360 degree or front–back fusion. The literature is replete with studies on the effects of fusion on clinical and radiographic outcomes but the studies differ by design, surgical indications, and techniques. Chou et al, in a systematic review of the literature, identified 18 randomized trials evaluating the outcomes of surgery for lumbar DDD.³ Four level I randomized, controlled trials (RCTs) comparing surgery (fusion) versus conservative treatment have been published,^4–7 two of which enrolled more than 100 patients.

Fritzell et al7 evaluated the role of lumbar fusion in a prospective, randomized clinical trial. The results published from the Swedish Lumbar Spine Study Group included 2-year follow-up with independent review of 294 patients from 19 centers. All patients were diagnosed with lumbar diskogenic pain and were randomized into a continued conservative treatment arm (n = 72) or one of three surgical arms (n = 222) including posterolateral noninstrumented fusion, instrumented PLIF, or circumferential fusion. The results revealed that the surgically treated patients experienced significant improvement based on Oswestry Disability Index (ODI) and patient satisfaction scales, and return to work that was greater than for those in the nonoperative treatment group. Forty-six percent of those randomized to surgery achieved good to excellent results compared with 18% of those randomized to nonoperative treatment (p < 0.0001). The authors concluded that surgery for CLBP provided symptomatic and functional improvement given available diagnostic tools, but no conclusion could be made as to which surgical treatment was more effective. In addition, they noted diminishing favorable results with time.

Brox et al⁴ reported on an RCT comparing posterolateral fusion to cognitive intervention and exercise for the treatment of CLBP. Although the sample size in this study was much smaller than that in the Fritzell et al study (n = 64 patients), and follow-up was only 1 year compared with 2 years, there was no reported difference in clinical pain outcome or disability (ODI). The authors concluded that aggressive nonoperative treatment was as effective as surgery in managing these patients. Inconsistency between these trials and others may be related to the different interventions utilized in the nonoperative groups.

These, among numerous other studies, have fueled the debate on the optimal treatment for the complex clinical syndrome of CLBP, although fusion is likely still considered the gold standard of care for the patient who has failed nonoperative management. It is also widely acknowledged, however, that there are numerous shortcomings of spinal fusion, including variability of pain relief, loss of segmental motion, risk of pseudarthrosis, malalignment of segmental balance, donor site morbidity when autograft is harvested from the iliac crest, and adjacent segment degeneration. In an attempt to address these potential problems, alternatives to fusion and maintaining motion in the lumbar spine degenerative segment (s) have been investigated over the past 30 years. Total disk replacement (TDR) has several theoretical goals including restoring or maintaining motion at the index level of surgery and/or restoring or maintaining motion at the adjacent segments of the lumbar spine. Whether TDR actually accomplishes these goals remains controversial.

There are numerous designs of TDR on the market and under investigation today. Conceptual and completed designs have been evaluated and used in clinical settings outside of the United States for years. Critics of TDR point to the inferior results reported in the 1980s and 1990s as a deterrent to further use of these devices.^8–10 However, numerous iterations and improvements in design of the implants (sizes, angulations, coating surfaces), insertion instrumentation, and imaging have led to increased usage. In addition, refined indications have encouraged better results. The Charité artificial disk replacement (Depuy Spine, Raynham, MA) was the first TDR to gain approval by the U.S. Food and Drug Administration (FDA) in 2004. It is indicated for single-level DDD at L4–L5 or L5–S1 in a patient with CLBP with appropriate clinical and radiographic characteristics. The ProDisc-L TDR (Synthes, Paoli, PA) is also FDA-approved for use in the lumbar spine and is indicated for one- or two-level DDD from L3 to S1. The Maverick disk (Medtronic, Memphis, TN) and Flexicore TDR (Stryker Spine, Allendale, NJ) are others also currently being evaluated in IDE trials.

Although there are theoretical advantages of TDR over fusion for CLBP secondary to lumbar disk degeneration, a significant question remains whether these replacements are clinically relevant and applicable. To address these questions, a comprehensive literature review of prognostic, therapeutic, and diagnostic studies was performed to assess the best available published evidence for the following questions:

• Does TDR provide more consistent or better improvement of pain and function than fusion following surgery for lumbar DDD in the CLBP patient?

• Does TDR maintain or restore motion to the index level of surgery?

• Does TDR maintain or restore facet integrity at the index level of surgery?

• Does TDR maintain or restore motion to the adjacent segments of the lumbar spine and does this translate into less adjacent level problems than with fusion?

Numerous challenges to this effort exist when comparing outcomes and complications of the two differing approaches. Although indicated for the same clinical problem of lumbar DDD, the evaluation of index and adjacent-level complications are different for each technique (TDR vs fusion). For example, index-level progressive facet joint degeneration would obviously not be considered with fusion but is a concern with TDR. Other challenges when evaluating the literature and comparing these two conceptual approaches to lumbar DDD include the lack of long-term follow-up with fusion greater than 2 years, whereas TDR must be followed much longer than that to evaluate its proposed goals. Finally, donor site morbidity and pseudarthrosis are clearly not a risk with TDR, although this can unquestionably affect the clinical outcomes of a fusion.

Does TDR Provide More Consistent or Better Improvement of Pain and Function than Fusion following Surgery for Lumbar DDD in the CLBP Patient?

The clinical outcomes of TDR have been assessed with very few level I and II studies and mostly level III publications. Two RCTs have been published comparing TDR with fusion for lumbar DDD, but no studies have been reported comparing it to nonoperative treatment.11,12

Level I Studies

Blumenthal et al¹¹ reported level I data on the early follow-up outcomes of the Charité TDR in the U.S. FDA IDE trial. This was the first RCT comparing TDR and fusion performed in the United States and was designed as a noninferiority study with 2:1 randomization with adequate power analysis. The clinical study consisted of 304 patients with single-level L4–L5 or L5–S1 DDD and CLBP with 205 patients placed into the experimental arm (TDR) and 99 patients in the control (fusion) arm without significant baseline clinical differences between the populations. (Seventy-one patients in the nonrandomized “training” cases were not reported.) The fusion procedure was an ALIF with the BAK cage (Zimmer Spine, Minneapolis, MN) including autograft from the iliac crest, which at the initiation of the study was the only FDA-approved grafting material at that time. Clinical success was defined as greater than or equal to a 15-point (or 25%) improvement in ODI versus baseline data, no evidence of device failure, absence of major complications, and maintenance or improvement of neurological status for both groups. At 12 months, the follow-up was 95.8% in the TDR and 94.2% in the fusion group, whereas at 24 months, the rate was 91.5% and 89.2%, respectively.

There were no differences between the treatments in reference to operative time, blood loss, or levels of surgery. The TDR patients were discharged from the hospital at an earlier time point (3.7 vs 4.2 days, p = 0.0039). Preoperative VAS (100-point scale) and ODI scores were equivalent while at 12 months statistically greater improvements were detected in the TDR group. At 24 months, the visual analogue scale (VAS) score reduction compared with preoperatively was 40.6 in the TDR group and 34.1 in the fusion group, and the ODI change was 48.5% and 42.4%, respectively, both of which were not statistically significant.

Zigler et al¹² reported the early 2-year outcomes of the ProDisc-L TDR in a U.S. FDA IDE RCT. The randomization was similar to the Charité study and consisted of 292 patients, nonblinded after the surgery, in a 2:1 fashion single-level TDR to fusion assignment. The control fusion group technique was an ALIF combined with posterolateral fusion and instrumentation. Two-year follow-up rate was 98.6% for the TDR and 97.1% for the fusion groups.

Baseline demographics were similar between the groups. The TDR group showed statistically significant less intraoperative blood loss, operative time, and length of hospitalization (p < 0.0001), which was not surprising given the circumferential surgical approach in the control group.

ODI success was defined as a >15% improvement from baseline. The 24-month ODI improvement was similar when considering the “at least 15% improvement” but if “at least 25% improvement” is considered, 69.1% of the TDR patients and 54.9% of fusion patients achieved these outcomes, which was statistically significant (p = 0.0396). At 24 months, the mean VAS improvement was 39 mm for the TDR and 32 mm for the fusion group (p = 0.08).

Level II Studies

Guyer et al¹³ have recently reported on the 5-year outcomes of the U.S. IDE trial on the Charité TDR. This is the first published long term, multicenter RCT comparing TDR and fusion. A potential limitation of this study was that of the 14 original centers enrolling patients into the study, eight sites eventually participated in the long-term follow-up data collection making this level II evidence, although, of those eight centers, each reported 100% follow-up. Two hundred and seventy-seven patients (233 randomized and 44 nonrandomized training cases) were eligible for 5-year follow-up, whereas only 160 actually did undergo eventual evaluation. The combined data consisted of 133 patients with 90 randomized to the Charité and 43 to fusion. This represented 57% of randomized eligible patients and 44% of the total IDE population.

The results reported showed a 57.8% success rate in the TDR and 51.2% in the fusion patients (p = 0.0359). No significant difference was detected in ODI, VAS pain scores, or Short Form-36 (SF-36) functional outcomes scales when comparing the two treatment groups. There were statistically significant findings in return to work criteria where 65.6% of TDR and 46.5% of fusion patients accomplished employment at 5-year follow-up (p = 0.0403). The authors concluded that, although no statistically significant differences in clinical outcome were identified, the results were consistent with the 2-year outcomes and confirmed the study design of noninferiority of the Charité TDR compared with ALIF and equivalent pain and functional outcomes.

Level III Studies

In 2005, Lemaire et al¹⁴ reported level III evidence on the long-term outcomes of TDR. They published a retrospective review of a consecutive, nonrandomized series of 107 patients with minimum 10-year follow-up (10 to 13.4 years) of which 100 were available for review. One hundred forty-seven Charité TDR implants were placed via the standard anterior retroperitoneal approach to the lumbar spine with 54 single-level, 45 two-level, and one three-level procedure. Clinical outcomes were assessed by a Modified Stauffer-Coventry scoring system, which revealed 62% excellent, 28% good, and 10% poor results. This scoring system was originally designed for evaluation of surgical outcomes for lumbar spinal stenosis and takes into consideration pain, medication usage, and resumption of same work activities. Employment data confirmed 91.5% of patients returned to their same or different job at long-term follow-up, whereas 63.2% where involved in a heavy labor defined occupation. The authors concluded that these data are consistent with outcomes from lumbar fusion and that TDR is a viable alternative to fusion.

David published in 2007 his long-term outcomes with level III evidence on the clinical effectiveness of TDR.15 In a retrospective chart review, he identified 108 patients treated with single-level Charité TDR at L4–L5 or L5–S1 with 106 available for mean follow-up at 13.2 years (10 to 16.8 years). Similarly, he utilized the Modified Stauffer-Coventry scoring system and reported 82.1% good/excellent results and 89.6% return to work including 77.8% in a heavy labor position. Although both of these level III studies confirmed clinical improvement with TDR, neither claimed to be superior to fusion.

Zindrick et al¹⁶ more recently have performed an extensive systematic review of the contemporary TDR literature in an attempt to identify factors affecting the clinical outcomes of these devices. They reviewed the literature from 1990 to 2007 and identified 76 clinical manuscripts of which 49 were excluded for the lack of relevant data and 27 were eventually reviewed. Thirteen evaluated the Charité and ProDisc-L TDR and one reported on the Maverick TDR. Most of these publications were level III and IV case-controlled series not having a comparison group and four level I studies. The reviews confirmed the lack of consistency in clinical outcomes, which often provide conflicting results.^17–22 For example, 10 level IV studies evaluated the role of multiple levels versus single-level implantations, six of which reported similar results. None of these compared the results with fusion. Other factors that could affect clinical outcome including the specified level of surgery and patient age did not reveal significant trends and again were not compared with fusion.

Summary of Data

There are limited long-term data on the effectiveness of TDR on clinical outcomes, specifically when comparing with fusion and no available studies comparing to nonoperative treatment (Table 25.1). The level II study recently published by Guyer et al¹³ is the first long-term 5-year RCT comparing the two procedures and, although limited by dropout rate, does provide some evidence of comparable clinical outcomes. When considering the two level III longer-term follow-up studies, it is fair to conclude that TDR, when technically performed well and with appropriate indications, results in comparable pain and functional improvement when compared with fusion. It is not possible to claim that TDR results in superior outcomes based on the literature. If clinical outcomes only are considered, the best available evidence suggests that there is fair evidence that TDR may be considered as an alternative to fusion (grade B recommendation).

Related posts:

Clearing Cervical Spine Injuries: MRI, Dynamic X-rays, CT Surgery for Axial Back Pain: ALIF versus PLIF or TLIF Avoidance of Cervical Adjacent Segment Disease Bone Morphogenetic Protein in the Setting of Infection: Indications, Risks, and Efficacy Vertebral Compression Fractures: Percutaneous Vertebral Augmentation Cervical Myelopathy: Meaning of Magnetic Resonance Imaging Signal Changes

Stay updated, free articles. Join our Telegram channel

Join