Results of a double-blinded randomised controlled trial indicate that anterior cervical disc arthroplasty (ACDA) does not lead to a superior outcome in comparison to anterior cervical discectomy with fusion (ACDF) or anterior cervical discectomy (ACD) alone. Subsequent investigation involving a larger patient cohort supports the “remarkable” nature of the similarity between outcomes of the three treatments, and even subgroup analysis “does not indicate that a certain type of patient would benefit more from one particular treatment strategy.”
“In contrast to what is generally believed,” the investigators comment, the Netherlands Cervical Kinematics, or NECK trial demonstrates “no significant differences” in clinical or radiological outcomes along single level ACDA, ACDF, and ACD. They conclude that all three options are “acceptable” forms of treatment and suggest that ACD without implanting an intervertebral device “may be a reasonable alternative” to ACDF or ACDA.
Writing in The Spine Journal, Carmen LA Vleggeert- Lankamp of Leiden-the Hague Spine Center (LUMC-HMC; Leiden/The Hague, The Netherlands) and colleagues found “no strong evidence” in favour of one of the three treatment strategies based on a two-year evaluation. Co-author Caroline MW Goedmakers (LUMC-HMC) presented a subgroup analysis based on the NECK trial and the Procon trial (conducted by Ronald Bartels and Roland Donk in Nijmegen, The Netherlands) combined, which found only “small differences” in treatment effect. These subsequent results were presented at the 46th annual meeting of the Cervical Spine Research Society in Scottsdale, USA (CSRS; 6–8 December, 2018).
Based on the available literature, Vleggeert-Lankamp and colleagues hypothesised that ACDA would lead to a “superior clinical outcome” in comparison to ACDF or ACD. In order to investigate this, 109 patients were randomised into one of three treatment groups: ACDA, ACDF with an intervertebral cage, and ACD without fusion. The authors measured clinical and radiological outcome by neck disability index (NDI), Visual Analogue Scale (VAS) neck pain, VAS arm pain, SF36, EQ-5D, patients’ self-reported perceived recovery, radiographic cervical curvature, and adjacent segment degeneration parameters at baseline until two years after surgery. BBraun Medical paid €298.84 to cover the costs for research nurses.
The authors conclude that “no statistical differences were demonstrated between the treatment groups.” Specifically, the NDI declined from 41 to 47 points at baseline to 19 ± 15 in the ACD group, 19 ± 18 in the ACDF group, and 20 ± 22 in the ACDA group after surgery. VAS arm and neck pain declined to half its baseline value and decreased below the critical value of 40mm. Quality of life, measured by the EQ-5D, increased in all three groups. Adjacent segment degeneration parameters were comparable in all three groups as well.
Vleggeert-Lankamp and colleagues comment on the potentially positive financial consequences of the study findings. Given that surgery with a prosthesis is “more expensive than tested alternatives,” the results that ACD is just as effective as both ACDA and ACDF may allow for savings to be made. They also mention a probable time saving element, as surgery to introduce a prosthesis is “longer and more challenging.” It has been demonstrated that this does not lead to more complications during or after surgery, however.
Since first described by Smith, Robinson, Cloward, Dereymaeker and Mulier in the 1950s, ACDF has been the “standard surgical treatment” for patients suffering from radiculopathy due to a herniated disc, Goedmakers comments. However, it is hypothesised that fusing the segments may increase the incidence of adjacent segment disease. In order to simulate the mobile characteristics of a cervical disc, the cervical intervertebral disc prosthesis was introduced.
Following the introduction of the prosthesis, the device has been the subject of several clinical and biomechanical studies. Trials comparing ACDF and prosthesis (ACDA) were widely performed but focused mainly on safety rather than superiority of the prosthesis.
Some years after the description of ACDF, anterior cervical discectomy without fusion (ACD) was reported. Although “very promising,” according to Goedmakers, it received less attention. The major comment was local kyphosis and thus ACDF became the “golden standard,” she comments. “We assume that this is the reason that all recent comparative studies focused on arthroplasty and ACDF, while ACD has not been given proper attention.” In the present study, however, the authors note that only two patients developed kyphosis.
In order to substantiate their results, the authors note that “larger studies with subgroup analyses are needed.” They also comment that the follow-up period of two years is a limitation of the present study. It is possible that adjacent level disease will occur after this period and that is has subsequent clinical effects. “We do however consider a follow-up period of two years an appropriate length of time to compare the initial clinical outcome of the three surgical methods.”
Talking to Spinal News International, Jack Zigler comments that: “While it is true that clinical outcomes are not significantly different between arthroplasty and ACDF or ACD [at least in the short term], reoperation rates in every one of the US IDE trials of cervical ADR versus fusion showed significantly fewer reoperations in the arthroplasty group, at every temporal data point.” He explains how “these were all individual Level I studies with prospective randomisation and high percentage multi-year follow-up. Several meta-analyses demonstrated this differential favouring ADR. IDE follow-up data was published for most implants at 7 years.”
Despite the “excellent methodology” of the NECK trial, he adds, “this level of evidence, generalised by thousands of geographically separated patients, hundreds of surgeons and site, with FDA oversight of methods and analysis, should be given stronger consideration than the 109 patients followed for only two years in the NECK trial.” He mentions that discectomy without fusion, for example, “may have equivalent two year outcomes to other reconstructive techniques, but is acknowledged to frequently results in later disc space collapse with recurrent foramenal narrowing and malalignment issues.”
Addressing the issue of cohort size at the CSRS annual meeting at the end of last year, co-author Goedmakers described a follow-up study in which data were collected from the 109 patients included in the NECK trial (LUMC; activC) and 142 patients from the Procon trial (RadboudUMC; Pro-disc) and combined, creating a total study population of 251 patients. The cohort was then divided into three treatment groups—ACD (83 patients); ACDF (83) and ACDA (85)—and was deemed large enough to allow for subgroup analysis.
In this larger patient cohort, the authors describe how the NDI decreased “comparably” in all treatment arms from 39.58–41.85 preoperatively to 14.62 ± 3.1 (ACD), 22.5 ± 3.6 (ACDF) and 22.7 ± 4.3 (ACDA) after two years. Treatment effect differences were 0.688 (CI -4.62 to 5.99; ACDF vs ACD), 1.113 (CI -4.77 to 6.99; ACDA vs ACD) and 0.424 (CI -5.79 to 6.64 ACDA vs ACDF) on a 100-point scale. In all three comparisons the treatment effect never exceeded the Minimal Clinical Important Difference (MCID) of 20 points. They found that results for the secondary outcome scales were also comparable. Subgroup analysis for gender, age, BMI, smoking and disc height at baseline demonstrate small differences in treatment effect without exceeding 20-point MCID borders, the investigators remark.
Goedmakers and colleagues describe how variables were analysed at baseline and at two years postoperatively. Subgroup analysis was performed for gender, age ≤40 and >40, BMI ≤30 and >30, smoking/non-smoking and disc height at baseline. To account for the correlation between repeated measurements Generalised Estimating Equations (GEE) were used. Study type was added as a fixed effect to correct for differences between the two studies.
Both the NECK trial and the Procon trial included patients with radicular signs and symptoms in one or both arms due to a single level cervical intervertebral disc herniation and/or an osteophyte in accordance with MRI findings. Both studies chose Neck Disability Index as the primary outcome variable, and subscales of the SF36 and McGill pain score as secondary outcome parameters.
The numbers of patients in RCTs are generally small, the investigators note, and therefore the results of two RCTs with comparable set up were combined. Both RCTs also included a group of patients in which no intervertebral device was implanted (ACD). As a result of the larger sample size in the combined dataset, the authors remark, subgroup analysis for BMI, age, gender and smoking could be performed.
Goedmakers comments: “It is possible that adjacent level disease will occur after the period of two years and that this would have subsequent clinical outcome effects. However, the Procon eight-year follow-up results do not demonstrate this tendency.” She adds that further to the CSRS presentations, data of longer follow-up periods will have to be collected and “critically evaluated” in order to verify if the conclusions of the NECK trial remain unchanged in the long term.