ABSTRACT
Biomechanical studies have demonstrated that cervical fusion results in increased motion and intradiscal pressures at adjacent levels. Cervical disc arthroplasty (CDA) is an alternative treatment for cervical radiculopathy and myelopathy resulting from degenerative disc disease. By maintaining segmental motion, surgeons hope to avoid some of the primary drawbacks of anterior cervical discectomy and fusion (ACDF), such as pseudoarthrosis and adjacent segment disease. First introduced in the 1960s, CDA has evolved over the years with changes to implant geometry and materials. Early devices produced suboptimal outcomes, but more recent generations of implants have shown long-term outcomes rivaling or even surpassing those of ACDF. In this article, the rationale for CDA as well as the history of such devices is reviewed.
INTRODUCTION
Anterior cervical discectomy and fusion (ACDF), first described in 1958 by Smith and Robinson,1 has been widely adopted as a standard treatment for cervical disc herniations unresponsive to conservative treatment as well as for cervical spondylotic radiculopathy and/or myelopathy. With the exception of the addition of anterior plate fixation and biologic substitutes, the procedure has remained largely unchanged, in part due to its reproducible results and good long-term clinical outcomes.2 ACDF is not without drawbacks, however, and efforts to address its limitations have led to the development of cervical disc arthroplasty (CDA). The rationale for motion preservation and the history of CDA will be expanded upon herein.
Rationale for Motion Preservation
Structural autologous iliac crest bone graft (ICBG) was used in the original description of the ACDF procedure and had traditionally been considered the gold standard.1 In a recent survey of over 5000 spine surgeons worldwide, however, only 20% reported using ICBG autograft in their routine single-level ACDFs.3 This is likely due to the drawbacks associated with this graft choice. Besides the additional operative time necessary to harvest structural autologous ICBG, patients may experience complications including persistent pain, neurologic injuries, seromas, hematomas, infection, vascular injuries, and pelvic fractures.4 Chronic pain is a commonly reported complication of graft harvest, with as many as 16.5% of patients reporting more severe pain at the harvest location than at the primary surgical site 1 year after surgery.5
It is therefore not surprising that many surgeons have moved away from ICBG. Structural allograft and cages made of polyether ether ketone (PEEK) and/or metals are now more prevalent in ACDF.3 However, these graft choices lack the biologic components of autologous ICBG. Pseudoarthrosis, a well-documented complication of ACDF, is commonly defined as failure of solid fusion at 1 year postoperatively and has been linked to poor clinical outcomes.6 A meta-analysis of prospective studies on the fusion rates associated with ACDF using anterior plate fixation reported a single-level pseudoarthrosis rate of 3.7%.7 Multiple studies have reported greater nonunion rates in multilevel procedures, and the incidence of pseudoarthrosis tends to increase with the number of levels included in the construct.8,9
In the early 2000s, surgeons experimented with recombinant human bone morphogenetic protein-2 (rhBMP2) use in the anterior cervical spine in an effort to increase fusion rates. Utilization in the anterior cervical spine declined after multiple studies reported significant and even life-threatening complications, including postoperative prevertebral swelling, dysphagia, and heterotopic bone formation.10,11 Efforts to improve the environment for fusion in PEEK and metal cages have more recently led to investigation into surface technologies, which may improve the fusion rate but also add cost.
Despite the above concerns regarding graft choice and pseudoarthrosis risk, most ACDF procedures produce fairly reliable symptomatic relief and go on to fusion.2,7 Patients undergoing single-level arthrodesis experience no decrease in global cervical range of motion (ROM), but fusing greater numbers of levels is associated with progressive loss of global motion (7.1° with 2 levels, 17.9° with 3 levels, and 22.1° with 4 levels).12 Furthermore, biomechanical analysis has demonstrated that fusion of even a single cervical motion segment results in motion compensation distributed throughout the remainder of the unfused levels, particularly at the segments immediately cranial and caudal to the surgical level.13 Intradiscal pressures in adjacent levels are thereby increased, which is believed to accelerate the degenerative process at these neighboring levels, contributing to increased radiographic adjacent segment degeneration and clinical adjacent segment disease.14
The incidence of radiographic adjacent segment degeneration after ACDF has been reported to be as high as 92% at 5-year follow-up,15 but questions arise as to the impact of fusion itself versus the natural history of the overall disease process. In order to address this issue, Matsumoto et al16 performed a prospective magnetic resonance imaging (MRI) study with 10-year follow-up comparing patients who underwent ACDF to healthy controls.16 They determined that while both groups demonstrated progression of disc degeneration over the course of 10 years, the incidence of progression at adjacent segments in ACDF patients was greater than that of control subjects, suggesting that fusion does in fact accelerate adjacent segment degeneration.
Clinical adjacent segment disease has been observed in many studies, most famously by Hilibrand et al,17 who followed 374 ACDF patients for up to 21 years. Authors observed that 25.6% of patients had symptomatic adjacent segment disease at the 10-year timepoint, and this developed at a relatively constant incidence of 2.9% annually over the course of these first 10 years.17 Over two-thirds of patients who developed clinical adjacent segment disease experienced symptoms of sufficient severity that they underwent additional surgery.
Given these limitations of ACDF, CDA has spurred interest as an alternative to fusion. Disc arthroplasty, like its fusion counterpart, allows for removal of the offending pathology with restoration of disc height. However, it features the additional benefit of maintained segmental motion, avoids the morbidity of ICBG harvest, and does not suffer from the issue of pseudoarthrosis.
These potential benefits of CDA are demonstrated in the long-term follow-up literature. Two-level CDA has now demonstrated superiority to 2-level ACDF in multiple Level 1 prospective, randomized controlled studies.18,19 Furthermore, a 2016 meta-analysis of 8 available randomized controlled trials reporting at least 4-year data on CDA versus ACDF found evidence to conclude that even single-level CDA produces significantly superior results in terms of overall success, neurologic outcome, neck pain improvement, surgery/implant-related serious adverse events, patient satisfaction, radiographic superior adjacent segment degeneration, and secondary surgeries (at both the index and adjacent levels).20
History of CDA
Although the widespread adoption of CDA may seem to be a relatively recent phenomenon, the designs of current implants are a result of decades of iteration. The first CDA was a stainless steel ball bearing device implanted by Ulf Fernstrom in 1966.21 However, this prosthesis ultimately fell out of favor due to high rates of complications, including adjacent-segment hypermobility, device migration, and subsidence.
The next generation of CDA devices did not appear for another 2 decades given the relative success of ACDF compared with the Fernstrom implant. Prompted by the popularity of lumbar arthroplasty in Europe in the 1980s, device developers and surgeons began experimenting with new CDA designs in the late 1980s and 1990s. B.H. Cummins at the Frenchay Hospital in Bristol, UK, introduced a CDA in 1989 that consisted of a stainless-steel ball-and-socket design secured with 2 anchoring screws. Initial clinical results of this metal-on-metal device were unsatisfactory—a report of 18 early patients revealed 4 cases of lost fixation, 1 case of instability, and persistent dysphagia in all 18 patients (possibly as a result of its high-profile design).22
The device was subsequently redesigned, with changes made to the device's profile, articulating surface geometry, and anterior locking mechanism. It was reintroduced as the Frenchay cervical disc and produced promising results in a 2002 pilot study.23 The device was later purchased by Medtronic, Inc, and renamed the Prestige disc. Results of a randomized clinical trial with 2-year follow-up comparing the outcomes of 276 patients undergoing arthroplasty with the Prestige ST disc (Figure 1) to 265 control patients who underwent ACDF for single level disc degeneration and radiculopathy was published in 2007.24 The CDA cohort exhibited a greater rate of neurological success, improved clinical outcomes (Neck Disability Index [NDI], neck pain, return to work), and a lower rate of secondary surgeries (both at the index and adjacent levels) than the ACDF group. There were no cases of implant migration or failure in the CDA group, and the device maintained average segmental sagittal angular motion of greater than 7° at final follow-up. In light of these data, the Prestige ST received US Food and Drug Administration (FDA) approval in 2007 for the treatment of cervical myelopathy and/or radiculopathy caused by disc herniation from C3 to C7 (Table).
Although the initial Prestige disc was made of stainless steel, the most recent iteration, the Prestige LP (low profile), is a titanium-ceramic composite that is MRI compatible and features a porous titanium coating to encourage bone in-growth (Figure 2). It received FDA approval as an alternative to ACDF for 1-level pathology from C3 to C7 in 2014 and for 2-level disease in 2016, based on large investigational device exemption (IDE) studies.25,26 Ten-year follow-up of the both the single- and 2-level study cohorts are now available—safety and effectiveness were maintained.27,28 In the single-level CDA cohort, 10.3% of patients had undergone a secondary surgery at the index level by 10 years, and 13.8% had a second surgery an at an adjacent level. In the 2-level study, the CDA group had significantly higher rates of overall success (80.4% versus 62.2%), NDI success (88.4% versus 76.5%), and neurological success (92.6% versus 86.1%). Improvements in NDI and neck pain scores were statistically superior for CDA as well. Patients in the Prestige LP group underwent significantly fewer secondary surgical procedures at the index-level (4.7% versus 17.6%) and at adjacent levels (9.0% versus 17.9%). All other study endpoints were at least noninferior for CDA as compared with ACDF. Furthermore, segmental angular ROM was maintained at both treated levels in the CDA group at final follow-up.
Vincent Bryan, an American neurosurgeon, designed the Bryan disc (Medtronic, Inc) in the early 1990s (Figure 3). This metal-on-plastic implant consists of a polyurethane core articulating between 2 titanium endplates with rough surfaces for bone ingrowth. The core is bathed in saline, which is contained by an outer polyurethane membrane to mimic synovial fluid, isolate wear debris, and prevent ingrowth of soft tissue. It is an unconstrained device held in the disc space by “press fit” rather than screw fixation. Ten-year outcomes from the original IDE cohorts have now been published with evidence of maintained motion, a greater rate of overall success composite endpoint (81.3% versus 66.3%, P = .005), and more improvement in NDI scores in patients (38.3 versus 31.1 decrease, P = .010) with single-level disease receiving the Bryan disc compared with those who underwent ACDF.29 A trend toward lower rates of secondary surgeries at the index level were observed with the Bryan disc compared with ACDF, but this did not attain statistical significance (9.7% versus 15.8%, P = .146).
The Prodisc-C (Centinel Spine), developed by surgeons Thierry Marnay and Rudolph Bertagnoli, was first implanted in Europe in 2002 (Figure 4). An FDA IDE study began in the United States the following year. The design of this device paralleled the principles of the ProDisc-L lumbar spine implant—it is a cobalt-chromium on ultra-high-molecular-weight polyethylene ball-and-socket design that features keels on each surface to anchor it to the vertebral endplates above and below. Seven-year data now available from the original FDA IDE study population demonstrate equivalent clinical outcomes to single-level ACDF with significantly lower probability of secondary surgery (18% in the ACDF group, 7% in the CDA group, P = .0099).30
There has subsequently been a proliferation in CDA devices. Over the years, they have generally become simpler to assemble and insert in order to facilitate surgeon acceptance and standardize outcomes. Currently available FDA-approved implants also include the SECURE-C by Globus Medical, Inc (Figure 5) and the Mobi-C by Zimmer Biomet (Figure 6), which is FDA-approved for 1- and 2-level use. Seven-year follow-up of the SECURE-C IDE cohorts reveals that this CDA implant produces statistically higher rates of overall success and patient satisfaction (96.0% versus 88.8%) than ACDF. Two-level CDA with the Mobi-C device has shown superiority with respect to 2-level ACDF in a large, multicenter randomized controlled study.19 Over the 2-year study window, patients with the Mobi-C experienced greater improvement in NDI scores at all timepoints, maintained their preoperative segmental ROM, and had a lower reoperation rate (11.4% versus 3.1%) than those with ACDF. The M6-C cervical disc (Figure 7), acquired by Orthofix in 2018 and approved for single-level use by the FDA in 2019, is unique in that it features a compressible artificial nucleus (made of polycarbonate urethane) and woven fiber annulus (made of polyethylene) intended to mimic the morphology and biomechanics of a native disc.31,32
The PCM disc, at one point offered by NuVasive, Inc, is not currently available on the market, although reasons for this have not been publicized (Figure 8). The Kineflex|C, developed by SpinalMotion, Inc, was a metal-on-metal (cobalt chrome molybdenum alloy) implant that completed a successful IDE study33 but ultimately did not achieve FDA approval due to concerns regarding metallosis given the bearing surface (Figure 9).
The currently available IDE study results are as follows: For single-level pathology, 2-year outcomes are available for the Prestige ST,24 Bryan,34 Prodisc-C,35 PCM,36 Prestige LP,25 SECURE-C,37 Mobi-C,38 and Kineflex|C33; 4-year data have been published for the Bryan39 and Prodisc-C40; 5-year data are available for the Prestige ST,41 Prodisc-C,42 PCM,43 and Mobi-C44; 7-year findings are published for the Prestige ST,45 Prodisc-C,30 Prestige LP,46 SECURE-C,47 and Mobi-C48; and 10-year outcomes are now available for the Bryan29 and Prestige LP27 implants. For 2-level surgery, 2-year data are available for the Prestige LP26 and Mobi-C19; 4-year49 and 5-year44,50 results have been published for Mobi-C; 7-year outcomes are available for the Prestige LP18 and Mobi-C48; and 10-year outcomes of the Prestige LP28 have been published.
Additional CDA designs are under development. The Simplify disc (Simplify Medical, Inc) has completed enrollment and is now in the data collection phase of both a 1- and 2-level IDE study (Figure 10).51,52 This disc uses PEEK endplates with a ceramic nucleus that produces less MRI artifact than the metallic endplates of other designs (Figure 11). There are other implants available outside of the United States that have not undergone the academically rigorous testing necessary for FDA approval.
CONCLUSION
Cervical disc replacement allows for the treatment of cervical disc disease while maintaining segmental motion. Advantages over traditional ACDF include preservation of ROM, avoidance of pseudoarthrosis, reduction in the incidence of adjacent segment disease, and a dramatic decrease in the need for secondary surgeries at both index and adjacent levels. Although the first CDA was performed over 50 years ago, the technology has gained momentum in the past 2 decades. The growing number of available implants, increasing mainstream adoption, and the availability of high-quality data demonstrating superior outcomes to ACDF suggest that CDA could ultimately become the new gold standard treatment for cervical disc disease.
Footnotes
Disclosures and COI: Dr. Derman declares the following relationships: Consultant agreements with Degen Medical, Orthofix, and Neo Spine. Teaching for Joimax. Research support from Aesculap and Orthofix. Dr. Zigler declares the following relationships: Consultant agreements with Aesculap, Centinel Spine, Medtronic, Orthofix, Simplify Medical. No outside sources of funding were utilized for the preparation of this manuscript.
- This manuscript is generously published free of charge by ISASS, the International Society for the Advancement of Spine Surgery. Copyright © 2020 ISASS.