Commercial

Artificial intervertebral disc insertion is an emerging technology intended for use in the cervical or lumbar spine to treat degenerative disc disease (DDD). DDD is a common cause of neck and/or low back pain. DDD is often defined as discogenic​ back pain with degeneration of the disc confirmed by individual history and radiographic studies. When conservative treatment for the disease (e.g., nonsteroidal anti-inflammatory drugs [NSAIDs], physical therapy) fails to relieve pain, a common surgical approach is spinal fusion. However, by surgically altering the biomechanics of the back, spinal fusion may also cause premature disc degeneration at adjacent levels, which results in increased pain and decreased range of motion (ROM). This is a notable concern for younger individuals. The artificial intervertebral disc insertion procedure (i.e., spinal arthroplasty or total disc replacement) was developed with the goal of avoiding the problems associated with spinal fusion surgery, which include a decreased ROM.

Artificial intervertebral disc insertion is intended to preserve ROM at the operative level once the diseased spinal disc is surgically removed, and an artificial disc is inserted between the adjoining vertebrae. The artificial disc generally consists of two metal endplates and a central, free component that moves within the disc space during spinal motion. Theoretically, the artificial intervertebral disc insertion procedure is designed to reduce or eliminate back pain and restore disc height while maintaining spinal curvature, flexibility, and load bearing. However, there are concerns that the use of an artificial disc increases the potential for implant failure due to device fracture, dislocation or wear, bone implant interface failure due to subsidence (i.e., sinking or settling in a bone), dislocation migration, or vertebral body fracture, and host response to the implant, which may include adverse events such as osteolysis, heterotopic ossification (HO), and pseudotumor formation.

Clinical studies have indicated that successful outcomes are typically only achieved by well-trained surgeons. This underscores the difficulty associated with the artificial intervertebral disc insertion procedure, especially when compared to spinal fusion, which is often considered the established standard for the treatment of DDD and low back pain. Additionally, there remain concerns about the need for revision surgery following artificial disc insertion, which often requires the disc to be removed. It is considered to be more complicated than revision surgery following spinal fusion due to the need for the abdomen to be opened, which results in severe blood loss.

CERVICAL ARTIFICIAL DISC REPLACEMENT

Cervical DDD is typically caused by spinal spondylosis (i.e., degenerative osteoarthritis), which causes intervertebral disc deterioration of the cervical spine. Symptoms may include radicular arm pain, weakness, and paresthesias (i.e., abnormal sensation). The prevalence of cervical DDD that is secondary to spinal spondylosis increases with age, with an estimated 60% of individuals older than 40 years having radiographic evidence of cervical DDD. By 65 years of age, approximately 95% of men and 70% of women have at least one degenerative change evidenced by radiography.

Anterior cervical discectomy and fusion (ACDF) is considered the definitive surgical treatment for symptomatic cervical DDD, and the resolution of pain and neurologic symptoms has been demonstrated in 80% to 100% of individuals undergoing ACDF. However, cervical artificial disc replacement (CADR) has been proposed as an alternative to ACDF for the treatment of individuals with neck pain, arm pain, nerve irritation, and/or spinal cord irritation due to cervical DDD that has not improved with at least 6 weeks of conservative management alone. It is hypothesized that CADR has the potential to reduce the risk of adjacent level DDD above or below a fusion site and to preserve ROM. However, while the intended patient populations for CADR and ACDF are similar, CADR should be limited to the treatment of one cervical spine level in individuals with DDD, without cervical anatomical deformities or compromised vertebral bodies at the treated level (e.g., ankylosing spondylitis, rheumatoid arthritis). Additionally, individuals should not have had prior spinal fusion at an adjacent cervical level, which may compromise the effectiveness of CADR at the treated level.

The FDA has approved several artificial disc devices for the treatment of DDD of the cervical spine including, but not limited to, ProDisc®-C (Synthes Spine, Inc., West Chester, PA), Prestige® ST Cervical Disc System (Medtronic Sofamor Danek, Memphis, TN), BRYAN® Cervical Disc (Medtronic Sofamor Danek, Memphis, TN), PCM Cervical (NuVasive, Inc., San Diego, CA), Secure®-C (Globus Medical, Inc., Audubon, PA), and Mobi-C® (LDR Spine USA, Inc., Austin, TX). Due to concerns about the long-term safety and effectiveness of these devices, postapproval studies were required to assess long-term data. These devices are indicated for spinal arthroplasty at one cervical spine level from C3 to C7 in skeletally mature individuals with DDD. There are multiple ongoing clinical trials that assess artificial intervertebral disc devices for the use in the cervical spine. The current FDA approvals were based on randomized clinical trials using a noninferiority trial design in which cervical artificial intervertebral disc insertion was compared to a type of spinal fusion surgery. The available published peer-reviewed literature for Secure®-C, if any, is limited. In August 2013, the Mobi-C® cervical disc received FDA approval for use in both one-level and two-level CADR.

The purpose of a noninferiority trial is to show that the experimental treatment is not clinically worse than (i.e., inferior to) the control treatment by a pre-specified margin, referred to as the noninferiority margin. A noninferiority trial is built on the assumption that the control has been proven effective. A superiority trial is a first choice for randomized controlled trial (RCT) trial design because it provides direct evidence of effect. In such a trial, the new treatment or device demonstrates a superior clinical result. A noninferiority trial should be a last choice for trial design because it offers indirect evidence of a treatment or device's effectiveness. These studies lack robust effectiveness data because decisions on degree of inferiority margins can be complex or biased, appropriate sample size can be difficult, and the results can be misleading and misinterpreted. The appropriateness of utilizing a noninferiority trial design for artificial intervertebral disc insertion has been questioned because the control (i.e., spinal fusion) has not been adequately studied to determine which type of spinal fusion surgery (e.g., anterior lumbar interbody fusion, posterior fusion) is the most effective or whether spinal fusion surgery is more effective than conservative management alone.

PEER-REVIEWED LITERATURE
In a prospective randomized controlled trial, Nabhan et al. (2007) compared the safety and effectiveness of ProDisc®-C with ACDF. Thirty-three individuals with symptomatic soft disc herniation were randomly assigned to CADR or ACDF (control group). These individuals were followed for 24 weeks and had a mean age of 45 years. Outcome measurements included segmental ROM postoperatively and neck and arm pain. The authors noted that segmental ROM was significantly better for extension and right-sided bending in the CADR group compared with the ACDF group (P<0.05). However, there was a statistically significant difference between both groups for right-sided axial rotation. The authors concluded that while neck and arm pain results were comparable between the two groups, loss of cervical segmental ROM was significantly higher in the ACDF. The study is limited in its small sample size, short-term follow-up period, and lack of blinding.

In a prospective RCT, Heller et al. (2009) evaluated the safety and effectiveness of CADR with the BRYAN® cervical disc compared with spinal fusion. Two hundred and forty-two individuals received the BRYAN® cervical disc and 221 individuals underwent one-level ACDF (control group). Study participants completed clinical and radiographic follow-up for up to 24 months after surgery. NDI was statistically significantly better in the CADR group compared with the ACDF group (P=0.025). There was no statistically significant difference between the two groups with regard to revision surgery subsequent to the index procedure. Individuals who received the CADR returned to work nearly 2 weeks earlier than individuals undergoing fusion (P=0.015). The authors concluded that CADR is a viable alternative to ACDF in individuals with persistent symptomatic, single-level cervical DDD. The study is limited in its short-term follow-up period and lack of blinding.

In a prospective RCT, Murrey et al. (2009) compared the safety and effectiveness of ProDisc®-C with ACDF for the treatment of DDD at one level between C3 and C7. A noninferiority design with a 1:1 randomization of 269 individuals was used (106 ACDF; 103 ProDisc®-C). Outcome measurements included visual analog scale (VAS) pain and intensity scores, NDI, and the number of adverse events. Individuals were followed for up to 24 months, with similar demographics between the two groups. The authors noted that both the ACDF and CADR groups had statistically significantly better NDI and VAS scores (P<0.0001). However, there was no statistically significant difference between the two treatment groups. At 24 months postoperatively, CADR individuals achieved a more than or equal to 4 degrees of motion or maintained motion relative to preoperative baseline measurements. The authors concluded that ProDisc®-C was a safe and effective treatment option for individuals with disabling cervical radiculopathy due to DDD. The study is limited in its short-term follow-up period and lack of blinding.

In a noncomparative, prospective study, Beaurain et al. (2009) evaluated the safety and effectiveness of the Mobi-C® device in 76 individuals with DDD at one or more levels. CADR was performed after radiological confirmation (e.g. CT, MRI) and failure of conservative treatment. Individuals older than 65 years of age or with osteoporosis were excluded. About 86.5% of study participants had no previous cervical surgery and 12.2% had previous fusion. Approximately 88% of study participants were operated with Mobi-C® at one level, while 12% were operated at two levels. Outcome measurements included NDI and VAS pain scores, in addition to ROM measurements pre- and postoperatively. Complications and reoperation rates were also assessed. The authors reported on intermediate 2-year follow-up results and analyzed occurrences of HO. Ultimately, only 58 individuals were evaluated at 2-year follow-up (i.e., 24% loss to follow-up). Of these study participants, 72% (n=42) met the successful outcome definition of preoperative NDI superior or equal to 30%, with a decrease of 15 points or more at 2-year follow-up. Mean NDI and VAS scores for both the arm and neck were statistically significantly reduced and ROM was preserved at index levels, with 85.5% of the segments remaining mobile at 2-year follow-up. In 72% of the study participants, HO was responsible for fusion at 6 of the 76 total levels. The authors concluded that the intermediate results of CADR were encouraging and that further studies were needed to determine the preservation of adjacent levels. The study is limited in its relatively short-term follow-up period, high loss to follow-up, heterogeneous study population, and lack of a comparative control group.

In a prospective RCT, Garrido et al. (2010) evaluated the safety and effectiveness of BRYAN® cervical disc in CADR when compared with ACDF. Forty-seven individuals with single-level DDD were followed for 48 months. Outcome measurements included NDI, VAS neck and arm scores, the number of adverse events, and the number of revision surgeries. At 48 months, NDI improvement was higher for CADR when compared with ACDF. VAS neck and arm pain scores were higher for ACDF when compared with CADR. There were six revision surgeries in the ACDF group and only one revision surgery in the BRYAN® cervical disc group. Of the six revision surgeries for the ACDF group, three were for adjacent level DDD and one was for remote level DDD. The remaining two surgeries were performed on the same individual for pseudarthrosis (i.e., unresolved fracture). In the CADR group, the revision surgery was for adjacent level DDD. The authors concluded that at 48 months, CADR continued to compare favorably to ACDF and resulted in no degradation of functional outcomes. The study is limited in its small sample size and short-term follow-up period.

In an RCT, Delamarter et al. (2010) reported on the 4-year follow-up results of the individuals enrolled in the FDA's investigational device exemption (IDE) trial for the ProDisc®-C. The authors noted that ACDF may result in a higher risk for revision surgery. However, both CADR and ACDF groups show good clinical results at 4-year follow-up. Furthermore, they noted that the study is limited by the lower individual accountability at 48 months compared with at 24 months. They indicated that follow-up was ongoing. Therefore, the reported results should be considered preliminary, and no conclusions can be drawn until the entire study population has been evaluated. Large, long-term RCTs are needed to demonstrate the long-term efficacy and safety of this device.

In a prospective, non-blinded RCT, Burkus et al. (2010) evaluated the long-term clinical outcomes in individuals undergoing anterior cervical surgery with the Prestige® device. Five hundred and forty-one individuals with cervical DDD were randomly assigned to CADR or spinal fusion (control group). Outcome measurements included NDI scores, health surveys, and neck and arm pain scores. Study participants were followed for up to 60 months. Of the 541 initial individuals, 271 (144 CADR and 127 spinal fusion) completed follow-up at 5 years. The overall rates of maintenance or improvement in neurological status in the CADR group were 91.6%, 92.8%, and 95.0% at 24, 36, and 60 months, respectively, compared with 83.6%, 83.2%, and 88.9% in the control group (P=0.006, 0.004, and 0.051). While rates of maintenance or improvement in the CADR group were not statistically significantly different at 60 months, the authors noted that rates for revision surgery at adjacent levels trended lower in the CADR group (eight individuals with 11 surgeries) compared with those in the spinal fusion group (13 individuals with 16 surgeries). These differences were not statistically different (P=0.376). Regardless, the authors concluded that the Prestige® disc maintained improved clinical outcomes and segmental motion at 5-year follow-up. The study is limited in its high loss to follow-up, mid-term follow-up period, and lack of blinding. For the purposes of an implantable device, a 5-year follow-up is insufficient to assess durability, the biologic effects of wear, and the response of the prosthesis to its environment, particularly in the relatively young population in which this device is being used.

In a prospective comparative study, Peng et al. (2011) evaluated the safety and effectiveness of the Prestige® disc compared with ACDF. Twenty-one female and 19 male individuals with a mean age of 43.9 years were followed for a mean of 2.9 years. Of these individuals, 62.5% had one-level replacement, 52.5% had myelopathy, and 47.5% had radiculopathy. Outcome measurements included NDI, cervical ROM, and radiography. The authors noted significant improvement in the NDI from a mean of 42.2 preoperatively to 16.4 at 6 months and 15.2 at 2 years (P<0.05). There was no statistically significant difference in clinical outcomes between CADR and ACDF. The authors concluded that CADR resulted in improved clinical outcomes at 2-year follow-up and can restore segmental lordosis (i.e., inward curvature of the spinal column) and preserve segmental ROM up to 2 years postoperatively. The study is limited in its lack of randomization, small sample size, short-term follow-up period, and heterogeneous study population.

In a prospective study, Huppert et al. (2011) evaluated the safety and effectiveness of the Mobi-C® device at one level compared with multilevel CADR. A total of 231 individuals with cervical DDD were evaluated at 2-year follow-up, with 175 undergoing one-level surgery and 56 at  two levels or more. Outcome measurements included NDI, VAS pain scores, ROM, and participant satisfaction. Mean NDI and VAS scores for both groups were improved, although there was no statistically significant difference between the two groups. In the multilevel group, analgesic use was significantly higher and the occurrence of HO was significantly lower. There was no significant difference in participant satisfaction. Approximately 11% of individuals in the one-level group and 20% of individuals in the multilevel group required reoperation or had at least one complication. While the authors concluded that there was no statistically significant difference between one-level and multi-level CADR, they acknowledged that further studies were necessary to understand the impact of multilevel CADR, especially on adjacent segments. The study is limited in its relatively short-term follow-up period, high loss to follow-up, heterogeneous study population, and lack of comparisons with gold standard treatments.

In a retrospective study, Ding et al. (2012) evaluated the intermediate clinical and radiographic outcomes of CADR with BRYAN® cervical disc. Thirty-four individuals representing 38 discs underwent CADR and study participants were followed for an average of 49.4 months (32 to 69). Clinical and radiographic outcomes, adjacent segment degeneration, complications, and reoperations were determined. NDI and neck and arm VAS pain scores were all statistically significantly improved postoperatively (P<0.05), but there were no statistically significant differences between the different follow-up time points. Approximately 23% of adjacent levels displayed mild degeneration at last follow-up. No revision surgeries were performed. The authors concluded that the BRYAN® cervical disc resulted in successful clinical outcomes; however, adjacent segment degeneration was observed. The study is limited in its retrospective study design, small sample size, and lack of a comparative group.

In a prospective RCT, Zhang et al. (2012) evaluated the safety and effectiveness of the BRYAN® cervical disc during CADR with conventional ACDF (control group). A total of 120 individuals were randomly assigned to CADR (n=60) or ACDF (n=60) and were followed for 24 months. Both groups had similar demographics including ROM, NDI, and VAS pain scores for the neck and arm. The CADR group had a statistically significantly longer operation time than the ACDF group (P<0.001). ROM was maintained in the CADR group, but was reduced in the ACDF group. There was no statistically significant difference between the two groups with respect to NDI or VAS pain scores. One individual in the CADR group and four individuals in the ACDF group required revision surgery. The authors concluded that CADR yielded good clinical results while preserving ROM at the index level. The study is limited in its small sample size, short-term follow-up period, and lack of blinding.

In a prospective study, Wu et al. (2012) evaluated the safety and effectiveness of one-level CADR compared with two-level CADR in 87 consecutive individuals. Ultimately, data from 70 individuals, representing 98 levels, were obtained. There were 42 individuals in the one-level group and 28 individuals in the two-level group. Outcome measurements included NDI and VAS pain scores and study participants were followed for a mean of 46 months. While both groups had statistically significant improvements in NDI and VAS, there was no statistically significant difference between the two groups. HO was identified significantly more frequently in the two-level group than the one-level group (75% vs. 40.5%; P=0.009). The authors concluded that clinical outcomes of both one-level and two-level CADR were similar at 46-month follow-up. However, there was a statistically significantly higher rate of HO in individuals undergoing two-level CADR. The study is limited in its nonrandomized design, relatively short-term follow-up period, and lack of a comparative control group.

In a prospective RCT, Delamarter and Zigler (2012) reported on the rates of secondary surgical intervention at both the index and adjacent levels for individuals treated with CADR or ACDF. Two hundred and nine individuals were randomly assigned (103 CADR, 106 ACDF) and followed for 5 years. A secondary surgical intervention was defined as a reoperation. Of the study participants that were ultimately involved in the study, 5-year follow up rates were 72.7% for the CADR group (n=72) and 63.5% for the ACDF group (n=61). At 5-year follow-up, individuals who received the ProDisc®-C device had a statistically significant probability of not having a reoperation at the index and adjacent levels when compared with individuals who underwent ACDF (97.1% vs. 85.5%; P=0.0079). There were no reoperations performed for implant breakages or device failures for the CADR group. For the ACDF group, the most common reason for reoperation at the index level was pseudarthrosis. For both groups, the most common reason for reoperation at the adjacent level was recurrent neck and/or arm pain. For the CADR group, three individuals had reoperations. One individual had a secondary surgical intervention to address persistent pain and two individuals were treated at the index and adjacent levels for persistent pain and adjacent level degeneration (ALD). The device was removed in two individuals and the levels were converted to anterior fusions. One device was left intact with a posterior foraminotomy and fusion with stabilization. The authors concluded that CADR results in decreased reoperation rates when compared with ACDF. The study is limited in its midterm follow-up period and its high loss to follow-up, which may hinder the internal validity of the study.

In a follow-up publication of the same study, Zigler et al. (2013) evaluated the safety and effectiveness of CADR using the ProDisc®-C device when compared with ACDF. Outcome measurements included NDI, VAS neck/arm pain scores, a SF-36 health survey (a standardized questionnaire used to measure an individual's overall subjective health status), neurological examination, device success, adverse event occurrence, and patient satisfaction. At 5-year follow-up, the authors reported that individuals in the CADR group had statistically significantly less neck pain intensity and frequency. Both groups scored high VAS satisfaction scores at 5 years. There were no reports of device failures or implant migration with the ProDisc®-C device and individuals in the CADR Group maintained ROM at their index level. The authors concluded that at 5 years, CADR was a safe and effective treatment for single-level symptomatic cervical disc disease (CDD) with clinical outcomes that are comparable to ACDF. The study is limited in its midterm follow-up period and its high loss to follow-up.

In a prospective, randomized FDA IDE pivotal study, Davis et al. (2013) evaluated the safety and effectiveness of CADR with the Mobi-C® device compared with ACDF for the treatment of 2-level symptomatic DDD. The primary outcome measurement was a composite measure of success at 2-year follow-up. A total of 330 individuals were enrolled and randomly assigned (225 received CADR; 105 received ACDF). At 2-year follow-up, 3% of individuals were lost to follow-up. Individuals in both groups showed statistically significant improvements in NDI and VAS pain scores. However, individuals undergoing CADR experienced statistically significant greater improvements in success and NDI. The reoperation rate was significantly higher in the ACDF group at 11.4% compared with 3.1% in the CADR group. Grade IV HO was present in 11 individuals undergoing CADR (5%) at 2-year follow-up. HO was not present in individuals undergoing ACDF. The authors concluded that on average, CADR was safe and effective compared with ACDF for the treatment of two-level symptomatic DDD. The study is limited in its relatively short-term follow-up period and potential risk for publication bias.

In a prospective, single-site study of two randomized clinical trials, Hacker et al. (2013) evaluated the safety and effectiveness of CADR using the Prestige® and BRYAN® devices compared with ACDF for the treatment of 94 individuals with cervical DDD. Nineteen individuals received the Prestige® device and 28 received the BRYAN® device. Outcome measurements included radiographical and clinical data scheduled postoperatively at 12, 24, 48, and 60 months. Adverse events were assessed as well. Late complications were defined as 24 months following surgery, and very late complications were defined as 48 months following surgery. Adjacent segment disease (ASD) occurred at a similar rate for individuals undergoing both fusion and CADR. Five individuals in the CADR group returned for evaluation of neck and arm symptoms 48 after surgery. Of these, four had peridevice vertebral body bone loss and one had posterior device migration and presented with myelopathy. Three required revision surgery. The authors concluded that despite the similarity between CADR and ACDF, they are not equivalent procedures in regard to very late complications. The authors also noted that appropriate follow-up intervals for CADR have not yet been defined by clinical trials. Therefore, they suggested that significantly longer follow-up periods may be warranted for individuals undergoing CADR than those undergoing fusion.

In a prospective RCT, Coric et al. (2013) evaluated the safety and effectiveness of CADR with conventional ACDF (control group) in individuals with single-level cervical radiculopathy. The results of two separate prospective randomized IDE trials (BRYAN® and Kineflex®-C) were combined to evaluate outcomes. Primary clinical outcome measurements included NDI, VAS pain scores, and neurological examination. A total of 74 individuals were randomly assigned to CADR (n=41) or ACDF (n=33). Eighty-six percent of individuals (n=63) completed a minimum of 4 years follow-up. Average follow-up was 6 years (48 to 108 months). In both the CADR and ACDF groups, mean NDI scores had a statistically significant improvement at 6 weeks after surgery and remained statistically significantly improved throughout the minimum of 48 months (P<0.001). ROM in the CADR group was statistically significantly greater when compared with the ACDF group. There were a total of three reoperations at the index or adjacent levels in the CADR group and there was one reoperation in the ACDF group. There were no statistically significant differences in overall reoperation rates. The authors concluded that both CADR and ACDF groups showed excellent clinical outcomes that were maintained over 48 months. The study is limited in its small sample size, heterogeneous treatment arms (BRYAN® and Kineflex®-C), and mid-term follow-up period.

In a retrospective study, Malham et al. (2014) evaluated the safety and effectiveness of the ProDisc®-C device in one-level (n=19) or two-level CADR (n=5). Outcome measurements included NDI and VAS pain scores. Complication and revision surgery rates were also noted. Average follow-up was 7.7 years. All outcome measurements had a statistically significant improvement. There were no episodes of device migration or subsidence, with a mean ROM of 6.4 degrees. HO)was present in 37% of individuals (n=7). Radiographic ASD below the device developed in 21% of individuals (n=4), with three occurring in individuals who underwent two-level CADR. The authors concluded that CADR was a safe and effective procedure, although there was radiographic evidence of HO and ASD on follow-up. The study is limited in its small sample size, retrospective study design, heterogeneous population, and lack of a comparative control group.

In a retrospective study, Fay et al. (2014) evaluated the safety and effectiveness of CADR using the BRYAN® cervical disc (n=37) compared with ACDF (n=40) in two-level cervical DDD. Seventy-seven consecutive individuals underwent two-level surgery and were followed for approximately 40 months. Outcome measurements included NDI and VAS pain scores. There were statistically significant improvements in all both NDI and VAS pain scores, though there was no significant difference between the groups. The authors concluded that clinical outcomes of two-level ACDF and CADR were similar 40 months after surgery; however, further studies were needed to truly establish the safety and effectiveness of surgery to treat multi-level DDD. The study is limited in its retrospective study design, relatively small sample size, and short-term follow-up period.

In a prospective RCT, Hisey et al. (2014) evaluated the Mobi-C® device in CADR when compared with ACDF for treating single-level cervical DDD. A total of 245 individuals were treated (164 CADR; 81 ACDF) and followed for 24 months. The primary outcome measurement was overall success based on improvement in NDI, no subsequent surgical interventions, and no adverse events. Secondary outcomes included VAS assessing neck and arm pain, patient satisfaction, radiographic ROM, and adjacent level degeneration. Overall success rates were 73.6% for CADR and 65.3% for ACDF, which confirmed noninferiority (P<0.0025). Operative level ROM in the CADR was maintained throughout follow-up, and radiographic evidence of inferior adjacent segment degeneration was significantly greater with ACDF at 12 and 24 months (P<0.05). The authors concluded that CADR with the Mobi-C® was a safe and effective treatment for single-level disc degeneration, producing similar outcomes when compared with ACDF. The study is limited in its relatively short-term follow-up period and potential for publication bias.

In a comparative study, Hey et al. (2013) evaluated the role of hybrid CADR and ACDF in seven consecutive individuals. Outcome measurements included VAS, NDI, and complication rates, and individuals were followed for 2 years. Data from the seven individuals who underwent the hybrid procedure were compared with a retrospective random selection of another seven ACDF and seven CADR individuals. The authors noted that the individuals who underwent the hybrid procedure returned to work faster when compared to either ACDF or CADR (P=0.035). There were no significant differences in ROM or functional scores. The authors concluded that the hybrid procedure was comparable to ACDF and CADR in terms of safety and feasibility, although additional large RCTs were warranted. The study is limited in its extremely small sample size, retrospective matching, and short-term follow-up period.

In a retrospective study, Park et al. (2013) evaluated the intermediate-term clinical and radiologic outcomes of CADR with Mobi-C®. The study population consisted of 75 individuals with cervical disc herniation, representing 85 disc levels. Mean follow-up was 40 months, with a minimum follow-up of 24 months. Outcome measurements included neck and arm pain scores and NDI. Cervical overall lordosis, segmental lordosis, and ROM were evaluated up to 24 months postoperatively. The mean numeric rating scale scores and NDI scores decreased significantly over 24 months. This represented an overall success rate of 86.7% according to Odom criteria. Mean segmental lordosis and motion increased and then decreased until 24 months. HO occurred in 67 levels at 12 months postoperatively, increasing to 80 levels at 24 months. The authors concluded that intermediate follow-up of CADR using the Mobi-C® device showed good clinical outcomes, although there was a trend toward reduced alignment and motion at 24 months. The overall HO occurrence was 94.1% at 24 months. The study is limited in its retrospective study design, lack of a comparative control group, and its relatively midterm follow-up period.

In a literature review, Alvin and Mroz (2014) evaluated the available literature on CADR with Mobi-C®, with a focus on two-level device. Fifteen studies evaluating CADR with Mobi-C® were included in the review, with study design, sample size, length of follow-up, statistical analysis, quality-of-life outcomes, conflicts of interest, and complications being recorded. Only one study was a level 1B RCT, with all included studies concluded a noninferiority of one-level CADR with Mobi-C® when compared with ACDF. Only one study analyzed outcomes of one-level versus two-level CADR with Mobi-C® and another evaluating two-level CADR with Mobi-C® when compared with two-level ACDF. The authors noted that in comparison with other CADR devices, the Mobi-C® device was associated with higher rates of HO. They concluded that one-level CADR with Mobi-C® was noninferior, but not superior, to one-level ACDF for individuals with cervical DDD. Additionally, they noted that insufficient evidence exists for two-level CADR with Mobi-C® when compared to two-level ACDF and that while Davis et al. (2013) did conclude superiority of two-level CADR with Mobi-C®, there were questions about publication bias. Specifically, they noted that the HO rate in the Davis et al. (2013) paper was 4.9% and different significant from every other study included in the review (range: 27.7%–94.1%) The authors indicated that there was a need for unbiased, well-designed prospective studies with well-defined outcomes. The study is limited in the heterogeneous nature of the included studies.

In a follow-up to the initial prospective, randomized FDA IDE pivotal study, Davis et al. (2015) evaluated the noninferiority of two-level CADR using Mobi-C® (n=225) when compared to two-level ACDF (n=105) at 4-year follow-up. At 24 months, the follow-up rate was 98.2% for the CADR group and 94.3% for the ACDF group. At 48 months, the follow-up rate was 89% for CADR and 81.2% for ACDF. Outcome measurements included NDI scores, patient satisfaction, and overall success. Both groups demonstrated significant improvement in NDI score, VAS neck pain, and VAS arm pain from baseline, with Mobi-C® meeting the noninferiority margin. Subsequent testing for superiority showed that CADR individuals had significantly greater improvement than ACDF with respect to NDI. CADR also resulted in significantly greater improvement in VAS neck pain at 6 months postoperatively, but not at 12, 24, 36, or 48 months. Arm pain scores did not differ between the groups. The CADR group had lower reoperation rates when compared with ACDF. At 48 months, adjacent level degeneration was observed in 41.5% of CADR individuals and 85.9% of ACDF individuals among those with available radiographs. Clinically relevant HO was observed in 25.6% of CADR individuals. Post-hoc analysis of the data from Davis et al. (2013) and Davis et al. (2015) were reported by Bae et al. (2015). Comparisons between single-level and two-level CADR with Mobi-C® revealed no significant difference on clinical outcomes (NDI, VAS, Short-Form 12), major complication rates, or subsequent surgery rates (3% for one level and 4% for two level). Clinically relevant HO was observed in 23.8% of individuals who underwent one-level CADR and 25.7% of individuals who underwent two-level CADR.

In a literature review, Skovrlj et al. (2015) evaluated the current available literature regarding reoperations following CADR. The authors noted that with increasing numbers of individuals undergoing CADR and longer available follow-up data, complications related to the devices and/or aging spine are growing. The published rates of reoperation (mean, 1.0%; range, 0%–3.1%), revision (mean, 0.2%; range, 0%–0.5%), and removal (mean, 1.2%; range 0%–1.9%) following CADR were low and comparable to the published rates following ACDF. The authors indicated that there was minimal literature and no guidelines with respect to the approaches and techniques in revision and for the removal of implants following CADR. Additionally, they called for longer term follow-up studies to assess implant survivorship and its effect on revision and removal rates.

In 2020, Reinas et al. performed a retrospective analysis of clinical records, pre- and postoperative neutral/dynamic X-rays of individuals who underwent cervical disc arthroplasty (CDA) in multilevel CDD of two to four levels. The sagittal balance parameters (C2-7 and index angle, SVA), global and segmental ROM), neck and arm VAS, Odom criteria, reoperation rate, ASD, and willingness to undertake the same procedure again were evaluated. Rate of HO was studied for follow-up for more than 2 years. A subgroup analysis was performed regarding two-level versus three- or four-level arthroplasty. Thirty-two individuals were included with mean age of 46 years (range, 30–63). Seventy-seven cervical disc levels were treated with the same artificial disc. Twenty-one individuals were operated on two levels, nine in three levels, and two individuals in four levels. Postoperatively, there was a decrease in SVA and an increase in global and index ROM. Mean visual analog scores of neck (nVAS) and visual analog scores of arm (aVAS) were decreased. Two-level versus three- to four-level individual subgroups showed a lower SVA, a slight increase in global and index ROM. HO was present in 9.9% of disc levels operated, none of them with grade three or four. The authors concluded that multilevel CDA provides good clinical and radiological outcomes, preserving global and segmental cervical mobility, while having a beneficial effect on sagittal balance. These results were for two to four levels, making this technique a valuable option in selected individuals with cervical multilevel CDD.