Artificial Disc Replacement for Degenerative Disc Disease of the Cervical Spine: A Technology Assessment

Judith Walsh, MD, MPH



In This Article

Technology Assessment (TA)

TA Criterion 1: The technology must have final approval from the appropriate government regulatory bodies.

The Prestige® Cervical Disc (Medtronic Sofamor Danek, Memphis, TN) received FDA Premarket Approval (PMA) on July 16, 2007.

The ProDisc-C® (Synthes Spine Inc., West Chester, PA) received PMA clearance on December 17 2007.

The Bryan® Cervical Disc System (Medtronic Sofamor Danek, West Chester, PA) received PMA clearance on May 12, 2009.

TA Criterion 1 is met.

TA Criterion 2: The scientific evidence must permit conclusions concerning the effectiveness of the technology regarding health outcomes.

The Medline database, Cochrane clinical trials database, Cochrane reviews database and Database of Abstracts of Reviews of Effects (DARE) were searched using the key words cervical disc or cervical vertebrae or replacement or prosthesis or artificial disc and also with the term cervical arthroplasty. The search was performed for the period from 1966 to July, 2009. The bibliographies of systematic reviews and key articles were manually searched for additional references and references were requested form the device manufacturers. The abstracts of citations were reviewed for relevance and all potentially relevant articles were reviewed in full.

Inclusion criteria:

  • Study had to compare AIDA with spinal fusion

  • Study had to be a randomized trial

  • Study had to measure clinical outcomes

  • Included only humans

  • Published in English as a peer reviewed article

Studies were excluded if they only focused on non-clinical outcomes such as radiographic changes or surgical technique. They were also excluded if they were retrospective and/or if they were case series.

A total of 646 potentially relevant articles were identified. 609 were excluded for not addressing the research question. A total of 37 studies were evaluated. 27 were excluded. Reasons for exclusion included not being a randomized controlled trial (RCT), not reporting clinical outcomes or not comparing AIDA with spinal fusion. Of the remaining, seven were either duplicate publications or earlier publications from the same series. Of these, three randomized trials of FDA approved cervical disc devices that evaluated clinical outcomes were identified and are included in this review.

Figure 1.

Selection of studies for inclusion in review

Details of the three randomized trials and the outcomes measured are described in Table 1. A description of the clinical outcomes evaluated in the trials follows.

The primary clinical outcomes measured in all of these studies were the Neck Disability Index (NDI) and Overall Success. Additional outcomes include neurologic success, neck and arm pain, and Short Form 36 (SF-36) scores. The clinical outcomes are defined below.

The Neck Disability Index (NDI) is the primary outcome in these studies and is used to measure the effects of neck pain associated with activities of daily living. It includes ten questions that include neck pain intensity, personal care, reading, headaches, concentration, work, driving, sleeping and recreation. Total NDI score is calculated by adding the results and converting to a percentage from 0–100%. A lower score indicates less disability. NDI scores are compared pre and post-operatively to determine improvement. In addition, the change in the NDI at follow-up is also reported in some studies.

Neck Disability Index Success: Another measure of the NDI is the Neck Disability Index Success: Neck Disability Index Success is defined as a >15 point improvement in NDI from baseline. A 15 point improvement is chosen because it is thought to be of clinical significance.

Overall Success: Overall success is based on successful outcomes with NDI score (≥15 point improvement from pre to post operative score and maintenance or improvement in neurological status). In addition, a patient could not have suffered a serious implant associated or implantation procedure associated adverse event or have undergone a second surgery classified as a failure.

Neurologic Success: Neurologic Success was based on postoperative maintenance or improvement in condition compared with the preoperative status. Indicators included motor function, sensory function and deep tendon reflexes.

SF-36: TheSF-36 is a self-administered test used to assess general health status. It measures specific health concepts related to physical function and limitations, social functioning and health perceptions. There are two scores- the physical component summary (PCS) and the mental component summary (MCS). The PCS score is based on physical functioning, role physical bodily pain and general health subscales. The MCS score is composed of the vitality, social functioning, emotional and mental health scales.

Post Operative Arm Pain and Neck Pain: Neck and arm pain was measured post operatively. It was often measured using a Visual Analogue Score (VAS). Often the neck and arm pain measures were compared pre and post-operatively.

Level of Evidence: 1

TA Criterion 2 is met

TA Criterion 3: The technology must improve net health outcomes.

Each of the three randomized trials compared artificial cervical disc replacement to cervical fusion. In accordance with FDA regulations for the evaluation of new devices, each of the studies was compared to cervical fusion using a non-inferiority design. The goal, therefore, was to show that the new technology (cervical disc replacement) was not inferior to the established technology, and thus a one sided comparison was used. The secondary hypothesis for each of these trials was a superiority hypothesis and was designed to evaluate whether or not disc replacement might be superior to cervical fusion, and for this comparison a two-sided hypothesis was used.

The BRYAN cervical disc trial was published in 2009. This was a multicenter study where patients with single level cervical DDD were randomized to receive either the BRYAN cervical disc (n=290) or anterior cervical discectomy and fusion (n=292).[14] The main study hypothesis was that the outcomes from disc replacement would be at least equivalent to fusion. Study subjects were unblinded prior to surgery. Among the 290 subjects randomized to the cervical disc, 37 declined treatment and 12 converted to anterior fusion. The investigators report the remaining 241 subjects, plus one randomized to fusion who converted to the cervical disc, as "randomized" to the cervical disc. Similarly among the 292 subjects initially randomized to fusion, 80 declined treatment, one converted to the cervical disc, and two were not treated according to the FDA protocol. The investigators report the remaining 209 subjects, plus the twelve subjects randomized to the cervical disc who converted to anterior fusion, as "randomized" to the anterior fusion. The difference in the rates of declining treatment as randomized was highly significant (p<0.001). 465 Patients were followed for two years. Initially, investigators and patients were blinded to the procedure. However, post-operatively, the investigational group was treated with a two week course of non steroidal anti-inflammatory drugs and was allowed to resume nonstrenuous activities as they pleased. Because of these post operative differences, "further blinding was not practical or ethical." At 24 month follow-up there was a 91.6% retention rate.

The primary clinical outcomes are listed in Table 2. The primary outcome was "overall success" which included the primary efficacy and safety measures. Overall success is described above. The second primary outcome was the Neck Disability Score. Additional outcomes included arm and neck pain, and SF-36 PCS and MCS scores.

At 24 months, both groups had improvements in the clinical outcomes. Overall success in the intervention group was 82.6% compared with 72.7% in the fusion group (p=0.005). NDI scores were 16.2 in the intervention group and 19.2 in the control group (p=0.025). NDI success (defined as >15 point improvement in the NDI) was 86% in the intervention group versus 78.9% in the fusion group (p=0.001). Improvements in SF-36 scores were comparable, as were measures of neck and arm pain. Patients who received the artificial disc returned to work about two weeks earlier than those who had fusion.

Given that this was a non-inferiority trial, an as-treated analysis was the primary analysis (versus an intention to treat analysis). There were 12 patients in the study who were randomly assigned to receive the artificial disc but who received the control treatment because of anatomic or technique difficulties during the surgery. Another important limitation is that 117 patients were randomly assigned but declined participation once receiving the assigned treatment, many because of dissatisfaction with the assigned treatment.

Based on follow up of the measured clinical outcomes, the Bryan cervical disc appears to be safe and efficacious at two year follow-up.

The Prestige trial was published in 2007.[11] This was a multicenter trial in which 541 patients with single level cervical DDD were randomized to cervical disc arthroplasty with the Prestige disc (n=276) or cervical fusion (n=265). Participants were not blinded.

The main study hypothesis was that the outcomes from disc replacement would be at least equivalent to fusion. A secondary hypothesis was that disc replacement was superior to fusion. Participants were followed for two years.

The primary clinical outcomes are listed in Table 3. The primary outcome was "overall success" which included the primary efficacy and safety measures. Overall success is described above. The second primary outcome was the Neck Disability Index Score. Additional outcomes included arm and neck pain, and SF-36 PCS and MCS scores

Overall success at 24 months was actually higher in the intervention group than in the control group. The NDI scores and NDI success were not statistically significantly different. The rate of neurologic success was greater in the intervention group. SF-36 scores and neck pain were improved more in the intervention group. There were fewer secondary surgeries in the intervention group.

Based on two year follow up of the measured clinical outcomes, the Prestige cervical disc appears to be safe and efficacious at two year follow-up.

The ProDisc-C trial was published in 2009 and is the study upon which FDA approval was based.[13] This was a multi-center study at 13 clinical sites. 209 patients were randomized to receive either the ProDisc artificial disc replacement (n=103) or ACDF (n=106). All participants had single level cervical disc disease that was nonresponsive to conservative treatment. All participants were blinded until after the surgery.

The main study hypothesis was that the outcomes from disc replacement would be at least equivalent to fusion. A secondary hypothesis was that disc replacement was superior to fusion. 209 Patients were followed for two years.

The 24 month results of the ProDisc study are described in Table 4. The primary outcomes of overall success and Neck Disability Index Success were comparable in the two groups. Overall success was 72.3 in the intervention group versus 68.3% in the control group. The results of the secondary outcomes were comparable. Fewer secondary surgeries were required in the intervention group. Few patients in the disc replacement group required narcotics at follow-up (10.1% vs. 18.5%; p=0.073). At 24 month follow-up, safety and efficacy were comparable in those who had disc replacement compared with those who got fusion.

In all three studies, the rate of adverse events was not inferior to the rate in the fusion group and were often less than those seen in the fusion group. Adverse events included device related complications, surgical complications and the need for secondary surgeries Study follow-up was not long enough to determine the effect of disc replacement versus surgery on the rate of development of DDD in adjacent discs.

The results of the three trials showed that cervical disc replacement was not inferior to cervical fusion based on two year clinical outcomes. The clinical outcomes evaluated are largely relevant. The NDI is an established scale that measures quality of life as impacted by neck pain. Neck and arm pain are also clinically significant. Neurologic success means that overall neurologic status cannot decline after the surgery. The SF-36 measures physical and mental functioning both pre and post operatively. Overall success is a stringent outcome and includes improvement in the NDI of a clinically significant level, maintenance or improvement of neurologic status, and the requirement for no adverse events or additional surgeries. However, the impact after two years of follow-up is not known.

A critically important question is how the risks and benefits of disc replacement will balance over the long term. Although the two year follow-up data are promising, they do not provide information about the long term impact of these procedures. The theoretic advantage of a decrease in the development of adjacent disc DDD has not been shown over the long term.

Thus, given the lack of information about the long term clinical impact of artificial cervical disc replacement, whether or not the technology ultimately improves net health outcomes is not known.

TA Criterion 3 is not met.

TA Criterion 4: The technology must be as beneficial as any established alternatives.

Cervical fusion is the standard of care for the treatment of symptomatic cervical DDD. All three of the RCTs of artificial disc replacement appropriately compare disc replacement to the standard of care, cervical fusion.

At two year follow-up, cervical disc replacement appears to be comparable to and not inferior to cervical fusion. It is plausible that cervical disc replacement will reduce the risk of future adjacent segment DDD, but this has not yet been shown. In addition, the long term effects of cervical disc replacement, including any possible long term benefits or future complications are not yet known.

Thus, although in the short term (two year follow up), cervical disc replacement has been shown to be not inferior to cervical fusion, and some clinical outcomes have even been shown to be improved, the impact of cervical disc replacement after two years is not known. In fact, in the editor's comments on the ProDisc trial published in The Spine Journal in 2009, the editor comments, "Early data suggest disc arthroplasty has equivalent short-term outcomes in the setting cervical radiculopathy. Longer-term follow-up is needed as late failure of arthroplasty is a reasonable concern." Thus, it is not known whether or not the technology is as beneficial as any of the established alternatives.

TA Criterion 4 is not met.

TA Criterion 5: The improvement must be attainable outside of the investigational setting.

To date, three randomized multi-center clinical trials have compared artificial cervical replacement to cervical spinal fusion, but have looked only at 24 month follow-up outcomes. Long term clinical outcomes have not yet been evaluated in any setting. Since an improvement in long term clinical outcomes has not yet been demonstrated in the investigational setting, it cannot be attainable outside the investigational setting.

TA Criterion 5 is not met.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: