Artificial disc replacement and adjacent-segment pathology: 10-year outcomes of a randomized trial

View More View Less
  • 1 Department of Surgical Sciences, Uppsala University Hospital, Uppsala, Sweden;
  • | 2 Neuro-Orthopaedic Center, Jönköping, and Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden;
  • | 3 Department of Radiology, Uppsala University Hospital, Uppsala, Sweden; and
  • | 4 Department of Learning, Informations, Management and Ethics (LIME), Medical Management Center, Health Economics and Health Evaluation Research Group, Karolinska Institutet, Stockholm and Center for Spine Surgery in Stockholm, Sophiahemmets Sjukhus, Stockholm, Sweden
Free access

OBJECTIVE

Artificial disc replacement (ADR) is designed to preserve motion and thus protect against adjacent-segment pathology (ASP) and act as an alternative treatment to fusion surgery. The question remains, how well do ADR devices perform after 10 years of follow-up compared with fusion surgery in terms of patient satisfaction, sustainability, and protection against ASP?

METHODS

This was the 10-year follow-up study of 153 participants who underwent ADR or fusion surgery after anterior decompression due to cervical degenerative radiculopathy (ISRCTN registration no. 44347115). Scores on the Neck Disability Index (NDI), EQ-5D, and visual analog scale for neck and arm pain were obtained from the Swedish Spine Registry and analyzed using ANCOVA. Information about secondary surgical procedures was collected from medical records and presented as Kaplan-Meier curves. MRI and flexion-extension radiography were performed, and ASP was graded according to the Miyazaki classification system.

RESULTS

Ten participants were lost to follow-up, which left 143 participants (80 underwent ADR and 65 underwent anterior cervical discectomy and fusion). There were no differences between groups in terms of patient-reported outcome measures (10-year difference in NDI scores 1.7 points, 95% CI −5.1 to 8.5, p = 0.61). Nineteen (24%) participants in the ADR group compared with 9 (14%) in the fusion group underwent secondary surgical procedures. The higher reoperation rate of the ADR group was mainly due to 11 female participants with device loosening. The rates of reoperation due to ASP were similar between groups, which was confirmed with MRI assessment of ASP that also showed no differences between the groups (p = 0.21).

CONCLUSIONS

This was the first 10-year follow-up study to compare ADR with fusion surgery and to provide MRI information for the assessment of ASP. The authors found no benefit of ADR over fusion surgery after anterior decompression for cervical degenerative radiculopathy.

ABBREVIATIONS

ADR = artificial disc replacement; ASP = adjacent-segment pathology; CASP = clinical ASP; ITT = intention to treat; MCID = minimum clinically important difference; NDI = Neck Disability Index; NSAID = nonsteroidal antiinflammatory drug; PP = per protocol; PROM = patient-reported outcome measure; RASP = radiographic ASP; ROM = range of motion; VAS = visual analog scale.

OBJECTIVE

Artificial disc replacement (ADR) is designed to preserve motion and thus protect against adjacent-segment pathology (ASP) and act as an alternative treatment to fusion surgery. The question remains, how well do ADR devices perform after 10 years of follow-up compared with fusion surgery in terms of patient satisfaction, sustainability, and protection against ASP?

METHODS

This was the 10-year follow-up study of 153 participants who underwent ADR or fusion surgery after anterior decompression due to cervical degenerative radiculopathy (ISRCTN registration no. 44347115). Scores on the Neck Disability Index (NDI), EQ-5D, and visual analog scale for neck and arm pain were obtained from the Swedish Spine Registry and analyzed using ANCOVA. Information about secondary surgical procedures was collected from medical records and presented as Kaplan-Meier curves. MRI and flexion-extension radiography were performed, and ASP was graded according to the Miyazaki classification system.

RESULTS

Ten participants were lost to follow-up, which left 143 participants (80 underwent ADR and 65 underwent anterior cervical discectomy and fusion). There were no differences between groups in terms of patient-reported outcome measures (10-year difference in NDI scores 1.7 points, 95% CI −5.1 to 8.5, p = 0.61). Nineteen (24%) participants in the ADR group compared with 9 (14%) in the fusion group underwent secondary surgical procedures. The higher reoperation rate of the ADR group was mainly due to 11 female participants with device loosening. The rates of reoperation due to ASP were similar between groups, which was confirmed with MRI assessment of ASP that also showed no differences between the groups (p = 0.21).

CONCLUSIONS

This was the first 10-year follow-up study to compare ADR with fusion surgery and to provide MRI information for the assessment of ASP. The authors found no benefit of ADR over fusion surgery after anterior decompression for cervical degenerative radiculopathy.

In Brief

The objective was to investigate whether artificial disc replacement protects against adjacent-segment pathology at 10 years of follow-up in comparison with fusion surgery after anterior decompression due to cervical degenerative radiculopathy. Progression of degenerative changes on MRI and number of reoperations performed on adjacent segments were similar between groups. The authors concluded that the intrinsic disc aging processes may be more important for the progression of degenerative changes than fusion surgery.

Cervical radiculopathy is typically the consequence of degenerative disc disease resulting in compression, inflammation, or damage to a cervical nerve. Symptoms include sharp radiating pain in the arm, numbness, weakness, and loss of reflexes at a given cervical nerve distribution. Diagnosis is based on symptoms and findings on physical examination and supported by imaging results. Surgery is an option when conservative treatments fail (such as medication and physiotherapy). When anterior decompression of a cervical nerve is indicated, the decision to perform a reconstruction with fusion surgery or an artificial disc replacement (ADR) needs to be made.

Anterior cervical discectomy and fusion is the gold-standard surgical treatment for cervical radiculopathy, but there are concerns about loss of motion in the fused segment and possible acceleration of degeneration in the adjacent segments. ADR is expected to prevent the development of adjacent-segment pathology (ASP) associated with fusion by preserving segmental motion at the index level, thereby leading to better clinical outcomes.1

Several nonblinded investigational device exemption studies have shown the clinical noninferiority of ADR devices in comparison with fusion surgery.2 A multicenter randomized controlled trial that compared ADR with fusion surgery was conducted by our group and 10 years of follow-up data have been obtained. The results at 2 and 5 years of follow-up have been published.3,4 There were no differences in clinical outcomes between the groups after 2 and 5 years. Complications after ADR surgery occurred, including spontaneous fusion and loosening. Five-year postoperative follow-up, which included MRI evaluations, revealed no differences between the groups in terms of ASP.

The aim of this study was to determine whether ADR surgery resulted in better long-term clinical outcomes than fusion surgery after anterior decompression in patients with degenerative cervical radiculopathy and to investigate with MRI whether ADR surgery prevented ASP at 10 years. Because none of the larger open-label randomized controlled trials had long-term follow-up MRI data at 10 years postoperatively,5–8 there are still insufficient data to support the view that ADR devices prevent ASP.

Methods

Trial Design

Between 2007 and 2011, patients with cervical radiculopathy were randomly assigned to undergo reconstruction with either an unconstrained ADR using the Discover disc (DePuy Spine, Johnson & Johnson) or fusion using an autologous iliac crest graft and plate after decompression at three Swedish study centers. The results of this multicenter randomized clinical superiority trial were previously reported after 2 and 5 years of follow-up.3,4 The enrolled participants were 25 to 60 years old, and the inclusion criteria were symptoms of radiating arm pain for at least 3 months, correlative findings on MRI at 1 or 2 cervical levels, eligibility for both treatments, and the ability to understand and read Swedish.

Exclusion criteria were previous cervical spine surgery, > 2 cervical levels requiring treatment, severe facet arthropathy, symptoms or marked radiological signs of myelopathy, drug abuse, dementia or expected poor compliance, cervical malformation or marked instability, history of severe cervical trauma, pregnancy, rheumatoid arthritis, malignancy, active infection or other systemic disease, and known allergy to an implant material or nonsteroidal antiinflammatory drugs (NSAIDs).

All participating surgeons were highly experienced in performing both trial interventions and performed the surgery with a standard anterior approach, as described by Smith and Robinson.9 When the affected nerve was decompressed, the posterior longitudinal ligament of the affected side was removed, as well as the uncovertebral joint if needed. Both the participant and surgeon were blinded until decompression was completed and the instruments to perform both ADR and fusion surgery were prepared. In the fusion group, reconstruction was achieved with a tricortical bone graft from the iliac crest and stabilized with a titanium plate of the surgeon’s preference. The postoperative regimen was the same in both groups, except 10 days of NSAID treatment was administered to the participants in the ADR group to prevent heterotopic ossification. No collars or restrictions were used in either group.

Data Collection and Outcomes

The primary outcome was NDI score, which ranges from 0% to 100%, with higher scores indicating severe disability. NDI constitutes 10 items that measure daily activities, concentration abilities, and severity of neck pain. The minimum clinically important difference (MCID) is 15%–17%.10–12

The secondary outcomes were scores on EQ-5D,13 EQ-5D health dimension, and visual analog scale (VAS) for neck and arm pain.14 EQ-5D (MCID 0.24) ranges from −0.5 to 1, and higher scores reflect better quality of life.12 EQ-5D health ranges from 0 to 100, with higher scores indicating better health. VAS-neck (MCID 21) and VAS-arm (MCID 29) range from 0 to 100, and higher scores indicate more severe pain.14

The primary and secondary outcomes were reported by the participant before surgery and 1, 2, 5, and 10 years after surgery. The participants also completed a questionnaire about preoperative baseline characteristics that was readministered at 1, 2, 5, and 10 years.

At the 10-year follow-up evaluation, the participants were examined with plain flexion-extension radiography and MRI, as well as T1- and T2-weighted imaging in the sagittal and axial planes. The findings on these examinations were compared with those of the preoperative imaging examinations and those obtained at the 5-year follow-up examinations. We determined the range of motion (ROM) of the index level by measuring the Cobb angles of the functional spinal unit,15 cervical sagittal vertical axis, and disc degeneration of adjacent segments according to the 5-level grading scheme by Miyazaki.16 We defined motion as a 5° or greater difference between flexion and extension of one functional spinal unit.17 Classification was done by a neuroradiologist (F.M.). ASP was defined as clinical ASP (CASP) if the participant had symptoms and signs at the adjacent segment that were consistent with the degenerative changes on MRI.18 CASP was defined as severe if nonoperative treatment failed and the participant underwent secondary surgery. Radiographic ASP (RASP) was defined as degenerative changes at the adjacent segment on MRI without any reported symptoms from the participant.

Study Oversight

The trial was approved by the local Swedish ethics review boards and registered at ISRCTN (registration no. 44347115). All participants provided oral and written informed consent, and the data were reported in accordance with the trial protocol. The authors designed the trial and vouch for the completeness and accuracy of the data, and no institution or company had any role in the data analysis, preparation of the manuscript, or the decision to submit the manuscript for publication.

Statistical Analysis

With 80% power to detect superiority in the primary outcome of NDI score, we calculated that a minimum of 102 participants would be required with α = 0.05 and an MCID of at least 103,19,20 between treatment groups. The final sample population was 153 participants to allow for dropouts, crossovers, and other unexpected events.

Patient-reported outcome measures (PROMs) were primarily analyzed as intention to treat (ITT), and data from all randomized participants were included. A per protocol (PP) analysis was performed to consider the effect of crossovers. To account for potential bias from missing values, multiple imputation using chained equations was performed as implemented in R package MICE21 with 20 iterations. We used ANCOVA to compare the mean 10-year PROMs between groups, and we adjusted for the baseline values of the outcomes. The results are presented as mean differences with 95% CIs and p values. A positive mean difference corresponded to greater values for ADR (compared with fusion).

Complications and reoperations were analyzed for the available participants because these outcomes could not be imputed. Time to revision surgery was analyzed with Kaplan-Meier plots, with men and women presented separately.

The development of RASP, as measured according to ΔMRI grade between the preoperative and 10-year MRI examinations, was compared between the ADR ITT and fusion groups, the ADR PP and fusion groups, and the mobile ADR and fusion groups with the unpaired 2-sided t-test. All statistical analyses were performed with R version 4.0.5 (R Foundation for Statistical Computing).22

Results

We randomly assigned 153 participants to undergo ADR (n = 83) or fusion (n = 70). There were 78 women and 75 men, with a mean (range) age of 47 (31–62) years (Table 1). The mean (range) follow-up was 10.7 (9.8–12) years. Sixteen participants in the fusion group and 18 participants in the ADR group did not report NDI scores at the 10-year follow-up (Fig. 1). The involved cervical levels of the entire participant cohort were C4–5 for 4 participants (3%), C5–6 for 50 (33%), C6–7 for 50 (33%), C7–T1 for 1 (1%), C4–6 for 4 (3%), and C5–7 for 44 (29%).

TABLE 1.

Baseline characteristics of all patients

CharacteristicADR (n = 83)ACDF (n = 70)Total (n = 153)
Male42 (50.6) 33 (47.1)75 (49.0)
Age, yrs46.9 ± 6.8 47.0 ± 6.947.0 ± 6.9
Smoking26 (31.3)21 (30.0)47 (30.7)
BMI, kg/m225.9 ± 4.326.2 ± 4.126.1 ± 4.2
2-level surgery28 (33.7) 20 (28.6)48 (31.4)
Unemployed8 (9.6)10 (14.3)18 (11.8)
Sick leave/disability pension56 (67.5)44 (62.9)100 (65.4)
Retired1 (1.2)0 (0.0) 1 (0.7)
Manual labor72 (86.7)54 (77.1) 126 (82.4)
NDI63.9 ± 16.5 61.5 ± 14.562.8 ± 15.7
EQ-5D0.4 ± 0.3 0.5 ± 0.3 0.4 ± 0.3
 Health46.9 ± 21.2 44.3 ± 19.3 45.6 ± 20.3
VAS
 Neck57.6 ± 26.4 58.2 ± 23.1 57.9 ± 24.9
 Arm57.1 ± 27.5 56.9 ± 23.0 57.0 ± 25.4

ACDF = anterior cervical discectomy and fusion.

Values are shown as number (%) or mean ± SD.

FIG. 1.
FIG. 1.

Flowchart/CONSORT diagram showing patient enrollment.

Outcomes at 10 Years

ITT Analysis

Both groups showed improvement in the primary outcome of NDI score. The ADR group improved from 64.1 (95% CI 60.4–67.7) to 25.3 (95% CI 20.6–30.0), and the fusion group from 61.4 (95% CI 57.8–65.0) to 22.4 (95% CI 16.8–28.0) (Fig. 2). The mean difference between the groups was 1.7 (95% CI −5.1 to 8.5, p = 0.61) (Table 2).

FIG. 2.
FIG. 2.

ITT analysis of ADR versus fusion. Mean NDI (A), EQ-5D (B), VAS-neck (C), and VAS-arm (D) scores at each time point (baseline and 1 year, 2 years, 5 years, and 10 years of follow-up). Error bars correspond to 95% CIs.

TABLE 2.

ITT analysis of baseline and 10-year follow-up imputed data of patients treated with ADR versus ACDF*

PROMADRACDFMean Difference (95% CI); p Value
Baseline10 YrsBaseline10 Yrs
Mean (95% CI)Median (range)Mean (95% CI)Median (range)Mean (95% CI)Median (range)Mean (95% CI)Median (range)
NDI64.1 (60.4–67.7)64 (26–100)25.3 (20.6–30.0)20 (0–88)61.4 (57.8–65.0)62 (0–96)22.4 (16.8–28.0)16 (0–88)1.7 (−5.1 to 8.5); 0.61
EQ−5D0.37 (0.30–0.44)0.3 (−0.2 to 0.8)0.71 (0.63–0.78)0.8 (−0.2 to 1.0)0.46 (0.39–0.53)0.6 (−0.2 to 0.8)0.74 (0.66–0.81)0.8 (−0.2 to 1.0)−0.01 (−0.11 to 0.08); 0.82
 Health46.9 (42.1–51.7)50 (10–95)63.6 (57.4–69.8)70 (0–100)43.8 (39.2–48.4)40 (0–81)73.5 (67.2–79.9)80 (0–100)−10.8 (−19.4 to −2.9); 0.015
VAS
 Neck57.0 (51.1–62.9)61 (0–100)31.8 (25.5–38.1)25 (0–100)58.1 (52.4–63.9)62 (0–100)28.8 (22.0–35.7)22 (0–100)3.2 (−5.8 to 12.3); 0.48
 Arm57.1 (50.9–63.2)60 (0–100)27.9 (20.9–35.0)16 (0–92)56.9 (51.2–62.6)62 (0–100)21.9 (15.0–28.8)9 (0–92)6.0 (−3.5 to 15.5); 0.21

Results were adjusted for baseline values.

ANCOVA-adjusted estimates of the 10-year differences between groups are reported.

There were no mean differences between groups in terms of the secondary outcomes of EQ-5D (mean −0.01, 95% CI, −0.11 to 0.08, p = 0.82), VAS-neck (3.2, 95% CI −5.8 to 12.3, p = 0.48), or VAS-arm (6.0, 95% CI −3.5 to 15.5, p = 0.21) (Table 2).

PP Analysis

The results of the PP analysis were similar to those of the ITT analysis (Table 3).

TABLE 3.

PP analysis of baseline and 10-year follow-up imputed data of patients treated with ADR versus ACDF*

PROMADRACDFMean Difference (95% CI); p Value
Baseline10 YrsBaseline10 Yrs
Mean (95% CI)Median (range)Mean (95% CI)Median (range)Mean (95% CI)Median (range)Mean (95% CI)Median (range)
NDI62.9 (58.9–67.0)62 (26–100)22.2 (17.5–27.0)18 (0–88)61.4 (57.8–65.0)62 (0–96)22.4 (16.8–28.0)16 (0–88)−0.7 (−7.8 to 6.3); 0.84
EQ-5D0.37 (0.29–0.46)0.3 (−0.2 to 0.8)0.74 (0.67–0.82)0.8 (−0.2 to 1.0)0.46 (0.39–0.53)0.6 (−0.2 to 0.8)0.74 (0.66–0.81)0.8 (−0.2 to 1.0)0.02 (−0.07 to 0.12); 0.64
 Health47.7 (42.1–53.2)50 (10–95)66.9 (59.9–73.8)71 (0–100)43.8 (39.2–48.4)40 (0–81)73.5 (67.2–79.9)80 (0–100)−7.6 (−16.8 to 1.5); 0.10
VAS
 Neck57.0 (50.0–64.1)63 (0–100)30.4 (23.4–37.5)22 (0–100)58.1 (52.4–63.9)62 (0–100)28.8 (22.0–35.7)22 (0–100)1.9 (−7.7 to 11.4); 0.70
 Arm57.4 (50.4–64.3)60 (0–100)27.0 (19.2–34.7)15 (0–92)56.9 (51.2–62.6)62 (0–100)21.9 (15.0–28.8)9 (0–92)5.0 (−4.9 to 14.9); 0.32

Results were adjusted for baseline values.

ANCOVA-adjusted estimates of the 10-year differences between groups are reported.

Secondary Surgery

Nineteen participants in the ADR group and 9 in the fusion group underwent secondary surgery (Fig. 3). The most common reason for reoperation was loosening and subsidence (n = 14) (Table 4). Most reoperations happened before 5 years of follow-up (24 participants), and only 4 participants underwent secondary surgery between 5 and 10 years of follow-up.

FIG. 3.
FIG. 3.

Kaplan-Meier plots of reoperations in each group stratified by sex.

TABLE 4.

Reoperations in the women and men included in the ADR and fusion groups*

Reason for ReopADRFusion
WomenMenWomenMen
Loosening & subsidence114
ASP825
Restenosis at index-level root canal111
Pseudarthrosis3
All patients14527

Values are shown as number of reoperations.

There were more reoperations than patients because 3 patients underwent reoperations for both loosening and ASP. In addition, 4 patients underwent a third operation, one of whom underwent a third reoperation for both loosening and ASP.

Preserved ROM and Alignment

In the ADR group, ROM changed from 43° at baseline to 50° at 10 years. In the fusion group, ROM changed from 45° at baseline to 36° at 10 years. The differences in ROM between groups were nonsignificant at baseline (p = 0.38) but significant at 10 years (p = 0.001). Of 58 prostheses (41 participants in the ADR group), only 38 (65%) had preserved motion (≥ 5° mobility in flexion and extension) at the operated cervical level and 20 (35%) had spontaneously fused.

There were no differences between groups in terms of cervical sagittal vertical axis at baseline (28 mm for the ADR group and 27 mm for the fusion group) or 10 years (28 mm for the ADR group and 30 mm for the fusion group).

Incidence Rates of CASP and RASP

Eight participants in the ADR group and 7 in the fusion group underwent secondary surgery due to severe CASP. When the changes on MRI of the participants without symptoms were assessed, mean RASP progressed from grade 2.6 preoperatively to grade 3.8 at 10 years postoperatively in the ADR group compared with 2.7 to 3.6 in the fusion group (p = 0.21). When the ADR ITT and ADR PP participants were compared with the fusion group participants, and when participants with mobile ADRs were compared with the participants who underwent fusion, the differences remained nonsignificant (Table 5).

TABLE 5.

ASP at the levels above and below the index segment at the baseline and 10-year follow-up evaluations, according to the classification system of Miyazaki et al.16

CharacteristicFusionADR ITTp Value*ADR PPp ValueMobile ADRp Value
Preop MRI
 Segment above treated index segment 2.6 (0.8)2.7 (0.8)0.552.65 (0.83)0.692.52 (0.71)0.74
 Segment below treated index segment2.0 (0.8)2.0 (0.7)1.001.9 (0.7)0.531.84 (0.62)0.37
10-yr MRI
 Segment above treated index segment3.8 (0.5)3.6 (0.7)0.213.64 (0.69)0.343.55 (0.69)0.14
 Segment below treated index segment3.3 (0.8)3.1 (0.8)0.213.04 (0.79)0.093.07 (0.87)0.76
ΔMRI§
 Segment above treated index segment 1.2 (0.9)1.0 (0.9)0.211.02 (0.94)0.421.08 (0.95)0.64
 Segment below treated index segment1.4 (1.1)1.1 (0.8)0.231.12 (0.82)0.291.2 (0.86)0.55

Values are shown as mean (SD) unless indicated otherwise.

ADR ITT versus fusion is shown.

ADR PP versus fusion is shown.

Mobile ADR versus fusion is shown.

Δ indicates the difference between the 10-year value and the preoperative value.

Discussion

The main argument for the use of ADR devices is the potential to prevent the progression of ASP. This is the first randomized study with 10 years of follow-up MRI data to report no protection of the adjacent segments with ADR surgery compared with fusion surgery.

Biomechanical and finite element analysis studies23–25 have reported increased stress on adjacent segments when simulated fusion procedures are performed. The stress effect seems to be mitigated in ADR models.24,26,27 Hilibrand et al. reported an annual incidence rate of 2.9% for ASP after fusion and predicted that 25% of patients would have new pathology at an adjacent segment within 10 years after fusion.1

RASP progressed by 1–1.5 grades (according to the classification system of Miyazaki et al.16), with no significant difference between the ADR and fusion groups. A comparison between the ADR PP group and fusion group was performed to account for the bias of crossovers, i.e., participants whose arthroplasty was removed and replaced with instrumented fusion. We also compared the mobile ADR group and the fusion group to account for the bias of participants with spontaneous fusion in the ADR PP group. None of these analyses showed any differences in RASP. The numbers of participants with severe CASP who required reoperation were similar (8 participants in the ADR group and 7 in the fusion group).

Nonoperated, already degenerated adjacent segments may be at higher risk for accelerated and eventually symptomatic degeneration after surgery, and therefore one could argue that it is advisable to include these segments in the primary surgery, whether that is fusion or ADR surgery. On the other hand, degenerative changes often progress without symptoms, and therefore we do not advocate prophylactic surgery. It is important to use a surgical method that minimizes the risk of accelerated ASP, but it is clear from the findings of our study that ADR is no such method when compared with fusion. Less invasive methods, such as posterior foraminotomy performed with tissue-sparing techniques or anterior decompression that spares the disc, may be considered and studied as alternatives.

By preserving segmental motion, ADR was expected to decrease the incidence of ASP associated with fusion and lead to better clinical outcomes. However, spontaneous fusion occurred in 27% of ADR implants at 5 years of follow-up.28 In our study with 10 years of follow-up, 34% of the ADR implants had unintended fusion. Spontaneous fusion tends to occur within the first 5 years after surgery, and then the process seems to slow down. The prevalence of spontaneous fusion varies greatly between studies, and one explanation has been interobserver error.29 Prevalence of spontaneous fusion also differs according to the predetermined value for degree of motion. We set the cutoff value at 5° because the minimal detectable change in Cobb angle was reportedly 5°.17 There was no difference in ASP between the participants whose ADR implants had preserved mobility and those participants whose implants had severe restriction of movement.

Secondary surgery was needed in 19 participants in the ADR group and 9 participants in the fusion group. The higher reoperation rate in the ADR group was mainly due to loosening/subsidence. Most reoperations were performed before 5 years of follow-up, which indicates that the symptoms of loosening and ASP somewhat stabilize a few years after surgery and then slow down. In our study, we used an unconstrained ball-socket implant, Discover, which has the highest reoperation rate among various ADR implants according to Wahood et al.30 As discussed in the 5-year follow-up study,4 most of the participants affected by loosening and subsidence were women, which is consistent with other reports.31 Even though patients with osteoporosis were excluded from this study, the enrolled participants were not screened for bone density. To account for all crossovers, we performed an as-treated PP analysis (Table 3), which presented similar results as the ITT analysis and showed no differences between groups.

Patient satisfaction (NDI, EQ-5D, VAS-neck, and VAS-arm scores) did not differ between the ADR and fusion groups. That finding is consistent with previous studies with 5 years of follow-up.4,32 In a double-blind randomized controlled trial,33 however, patient satisfaction was worse in the ADR group (higher NDI scores and worse arm pain) at 2 years of follow-up. The 10-year follow-up study reported NDI values at the same level as the 3-month follow-up for both groups3 and demonstrated improvement compared with the 2- and 5-year follow-up studies. The NDI values indicated that even though degeneration of the cervical spine progresses with age, patients do not progress in terms of perceived disability.

The main strength of our study was that our radiological follow-up included both dynamic radiography and MRI. Segmental degeneration is best assessed with MRI, and consequently our study with 10 years of follow-up MRI data presents a more accurate evaluation of the existence, progression, and grade of RASP.

One limitation of our study was that participants were lost to follow-up at 10 years, which we interpreted as participation bias. Some participants who missed their MRI appointment communicated that they were pleased with their operation and were completely symptom free. In contrast to the content participants, participants with chronic pain as a result of failed surgery tended to be more eager to participate. Because there was no significant difference in dropouts between groups, we think it is unlikely that the dropouts affected the comparison of outcomes.

Conclusions

This is the first 10-year follow-up study to compare ADR and fusion surgery after anterior decompression for cervical degenerative radiculopathy and to provide MRI information for assessment of ASP. Even though PROMs were similar between the ADR and fusion surgery groups at the 10-year follow-up evaluation, there were more reoperations in the ADR group due to implant loosening. The rates of reoperation for ASP were similar between groups, and preserved motion did not prevent ASP as assessed on MRI.

Disclosures

DePuy Synthes provided an unrestricted grant for this study.

Author Contributions

Conception and design: MacDowall, Skeppholm, Olerud. Acquisition of data: Kontakis. Analysis and interpretation of data: MacDowall, Kontakis, Löfgren, Mosavi, Skeppholm, Olerud. Drafting the article: MacDowall, Kontakis, Marques. Critically revising the article: MacDowall, Marques, Löfgren, Skeppholm, Olerud. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: MacDowall. Statistical analysis: MacDowall, Kontakis. Administrative/technical/material support: Mosavi. Study supervision: MacDowall.

References

  • 1

    Hilibrand AS, Carlson GD, Palumbo MA, Jones PK, Bohlman HH. Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg Am. 1999;81(4):519528.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2

    Phillips FM, Geisler FH, Gilder KM, Reah C, Howell KM, McAfee PC. Long-term outcomes of the US FDA IDE prospective, randomized controlled clinical trial comparing PCM cervical disc arthroplasty with anterior cervical discectomy and fusion. Spine (Phila Pa 1976).2015;40(10):674683.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3

    Skeppholm M, Lindgren L, Henriques T, Vavruch L, Löfgren H, Olerud C. The Discover artificial disc replacement versus fusion in cervical radiculopathy—a randomized controlled outcome trial with 2-year follow-up. Spine J. 2015;15(6):12841294.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4

    MacDowall A, Canto Moreira N, Marques C, Skeppholm M, Lindhagen L, Robinson Y, et al. Artificial disc replacement versus fusion in patients with cervical degenerative disc disease and radiculopathy: a randomized controlled trial with 5-year outcomes. J Neurosurg Spine. 2019;30(3):323331.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Radcliff K, Davis RJ, Hisey MS, Nunley PD, Hoffman GA, Jackson RJ, et al. Long-term evaluation of cervical disc arthroplasty with the Mobi-C© Cervical Disc: a randomized, prospective, multicenter clinical trial with seven-year follow-up. Int J Spine Surg. 2017;11(4):31.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6

    Mehren C, Heider F, Siepe CJ, Zillner B, Kothe R, Korge A, Mayer HM. Clinical and radiological outcome at 10 years of follow-up after total cervical disc replacement. Eur Spine J. 2017;26(9):24412449.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Gornet MF, Lanman TH, Burkus JK, Dryer RF, McConnell JR, Hodges SD, Schranck FW. Two-level cervical disc arthroplasty versus anterior cervical discectomy and fusion: 10-year outcomes of a prospective, randomized investigational device exemption clinical trial. J Neurosurg Spine. 2019;31(4):508518.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8

    Lavelle WF, Riew KD, Levi AD, Florman JE. Ten-year outcomes of cervical disc replacement with the BRYAN cervical disc: results from a prospective, randomized, controlled clinical trial. Spine (Phila Pa 1976).2019;44(9):601608.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9

    Smith GW, Robinson RA. The treatment of certain cervical-spine disorders by anterior removal of the intervertebral disc and interbody fusion. J Bone Joint Surg Am. 1958;40A(3):607624.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Carreon LY, Glassman SD, Campbell MJ, Anderson PA. Neck Disability Index, short form-36 physical component summary, and pain scales for neck and arm pain: the minimum clinically important difference and substantial clinical benefit after cervical spine fusion. Spine J. 2010;10(6):469474.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Young IA, Cleland JA, Michener LA, Brown C. Reliability, construct validity, and responsiveness of the Neck Disability Index, patient-specific functional scale, and numeric pain rating scale in patients with cervical radiculopathy. Am J Phys Med Rehabil. 2010;89(10):831839.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Parker SL, Godil SS, Shau DN, Mendenhall SK, McGirt MJ. Assessment of the minimum clinically important difference in pain, disability, and quality of life after anterior cervical discectomy and fusion: clinical article. J Neurosurg Spine. 2013;18(2):154160.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Strömqvist B, Fritzell P, Hägg O, Jönsson B. The Swedish Spine Register: development, design and utility. Eur Spine J. 2009;18(suppl 3):294304.

  • 14

    MacDowall A, Skeppholm M, Robinson Y, Olerud C. Validation of the visual analog scale in the cervical spine. J Neurosurg Spine. 2018;28(3):227235.

  • 15

    Kim SW, Limson MA, Kim SB, Arbatin JJ, Chang KY, Park MS, et al. Comparison of radiographic changes after ACDF versus Bryan disc arthroplasty in single and bi-level cases. Eur Spine J. 2009;18(2):218231.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Miyazaki M, Hong SW, Yoon SH, Morishita Y, Wang JC. Reliability of a magnetic resonance imaging-based grading system for cervical intervertebral disc degeneration. J Spinal Disord Tech. 2008;21(4):288292.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Marques C, Granström E, MacDowall A, Moreira NC, Skeppholm M, Olerud C. Accuracy and reliability of X-ray measurements in the cervical spine. Asian Spine J. 2020;14(2):169176.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Anderson PA, Andersson GB, Arnold PM, Brodke DS, Brodt ED, Chapman JR, et al. Terminology. Spine (Phila Pa 1976).2012;37(22 suppl):S8S9.

  • 19

    MacDermid JC, Walton DM, Avery S, Blanchard A, Etruw E, McAlpine C, Goldsmith CH. Measurement properties of the neck disability index: a systematic review. J Orthop Sports Phys Ther. 2009;39(5):400417.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Pool JJ, Ostelo RW, Hoving JL, Bouter LM, de Vet HC. Minimal clinically important change of the Neck Disability Index and the Numerical Rating Scale for patients with neck pain. Spine (Phila Pa 1976).2007;32(26):30473051.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21

    White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30(4):377399.

  • 22

    R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing; 2018.Accessed October 6, 2021. https://www.r-project.org/

    • Search Google Scholar
    • Export Citation
  • 23

    Goel VK, Faizan A, Palepu V, Bhattacharya S. Parameters that effect spine biomechanics following cervical disc replacement. Eur Spine J. 2012;21(5 Suppl):S688S699.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Eck JC, Humphreys SC, Lim TH, Jeong ST, Kim JG, Hodges SD, et al. Biomechanical study on the effect of cervical spine fusion on adjacent-level intradiscal pressure and segmental motion. Spine (Phila Pa 1976).2002;27(22):24312434.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25

    Li XF, Jin LY, Liang CG, Yin HL, Song XX. Adjacent-level biomechanics after single-level anterior cervical interbody fusion with anchored zero-profile spacer versus cage-plate construct: a finite element study. BMC Surg. 2020;20(1):66.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Lou J, Li Y, Wang B, Meng Y, Gong Q, Liu H. Biomechanical evaluation of cervical disc replacement with a novel prosthesis based on the physiological curvature of endplate. J Orthop Surg Res. 2018;13(1):41.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Faizan A, Goel VK, Garfin SR, Bono CM, Serhan H, Biyani A, et al. Do design variations in the artificial disc influence cervical spine biomechanics? A finite element investigation. Eur Spine J. 2012;21(5 Suppl):S653S662.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Marques C, MacDowall A, Skeppholm M, Canto Moreira N, Olerud C. Unintended fusion in cervical artificial disk replacement: a prospective study on heterotopic ossification, progression, and clinical outcome, with 5-year follow-up. Eur Spine J. 2021;30(6):16621669.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Kong L, Ma Q, Meng F, Cao J, Yu K, Shen Y. The prevalence of heterotopic ossification among patients after cervical artificial disc replacement: a systematic review and meta-analysis. Medicine (Baltimore). 2017;96(24):e7163.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30

    Wahood W, Yolcu YU, Kerezoudis P, Goyal A, Alvi MA, Freedman BA, Bydon M. Artificial discs in cervical disc replacement: a meta-analysis for comparison of long-term outcomes. World Neurosurg. 2020;134:598613.e5.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31

    Hacker FM, Babcock RM, Hacker RJ. Very late complications of cervical arthroplasty: results of 2 controlled randomized prospective studies from a single investigator site. Spine (Phila Pa 1976).2013;38(26):22232226.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32

    Delamarter RB, Zigler J. Five-year reoperation rates, cervical total disc replacement versus fusion, results of a prospective randomized clinical trial. Spine (Phila Pa 1976).2013;38(9):711717.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33

    Sundseth J, Fredriksli OA, Kolstad F, Johnsen LG, Pripp AH, Andresen H, et al. The Norwegian Cervical Arthroplasty Trial (NORCAT): 2-year clinical outcome after single-level cervical arthroplasty versus fusion-a prospective, single-blinded, randomized, controlled multicenter study. Eur Spine J. 2017;26(4):12251235.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • View in gallery

    Flowchart/CONSORT diagram showing patient enrollment.

  • View in gallery

    ITT analysis of ADR versus fusion. Mean NDI (A), EQ-5D (B), VAS-neck (C), and VAS-arm (D) scores at each time point (baseline and 1 year, 2 years, 5 years, and 10 years of follow-up). Error bars correspond to 95% CIs.

  • View in gallery

    Kaplan-Meier plots of reoperations in each group stratified by sex.

  • 1

    Hilibrand AS, Carlson GD, Palumbo MA, Jones PK, Bohlman HH. Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg Am. 1999;81(4):519528.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2

    Phillips FM, Geisler FH, Gilder KM, Reah C, Howell KM, McAfee PC. Long-term outcomes of the US FDA IDE prospective, randomized controlled clinical trial comparing PCM cervical disc arthroplasty with anterior cervical discectomy and fusion. Spine (Phila Pa 1976).2015;40(10):674683.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3

    Skeppholm M, Lindgren L, Henriques T, Vavruch L, Löfgren H, Olerud C. The Discover artificial disc replacement versus fusion in cervical radiculopathy—a randomized controlled outcome trial with 2-year follow-up. Spine J. 2015;15(6):12841294.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4

    MacDowall A, Canto Moreira N, Marques C, Skeppholm M, Lindhagen L, Robinson Y, et al. Artificial disc replacement versus fusion in patients with cervical degenerative disc disease and radiculopathy: a randomized controlled trial with 5-year outcomes. J Neurosurg Spine. 2019;30(3):323331.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Radcliff K, Davis RJ, Hisey MS, Nunley PD, Hoffman GA, Jackson RJ, et al. Long-term evaluation of cervical disc arthroplasty with the Mobi-C© Cervical Disc: a randomized, prospective, multicenter clinical trial with seven-year follow-up. Int J Spine Surg. 2017;11(4):31.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6

    Mehren C, Heider F, Siepe CJ, Zillner B, Kothe R, Korge A, Mayer HM. Clinical and radiological outcome at 10 years of follow-up after total cervical disc replacement. Eur Spine J. 2017;26(9):24412449.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Gornet MF, Lanman TH, Burkus JK, Dryer RF, McConnell JR, Hodges SD, Schranck FW. Two-level cervical disc arthroplasty versus anterior cervical discectomy and fusion: 10-year outcomes of a prospective, randomized investigational device exemption clinical trial. J Neurosurg Spine. 2019;31(4):508518.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8

    Lavelle WF, Riew KD, Levi AD, Florman JE. Ten-year outcomes of cervical disc replacement with the BRYAN cervical disc: results from a prospective, randomized, controlled clinical trial. Spine (Phila Pa 1976).2019;44(9):601608.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9

    Smith GW, Robinson RA. The treatment of certain cervical-spine disorders by anterior removal of the intervertebral disc and interbody fusion. J Bone Joint Surg Am. 1958;40A(3):607624.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Carreon LY, Glassman SD, Campbell MJ, Anderson PA. Neck Disability Index, short form-36 physical component summary, and pain scales for neck and arm pain: the minimum clinically important difference and substantial clinical benefit after cervical spine fusion. Spine J. 2010;10(6):469474.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Young IA, Cleland JA, Michener LA, Brown C. Reliability, construct validity, and responsiveness of the Neck Disability Index, patient-specific functional scale, and numeric pain rating scale in patients with cervical radiculopathy. Am J Phys Med Rehabil. 2010;89(10):831839.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Parker SL, Godil SS, Shau DN, Mendenhall SK, McGirt MJ. Assessment of the minimum clinically important difference in pain, disability, and quality of life after anterior cervical discectomy and fusion: clinical article. J Neurosurg Spine. 2013;18(2):154160.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Strömqvist B, Fritzell P, Hägg O, Jönsson B. The Swedish Spine Register: development, design and utility. Eur Spine J. 2009;18(suppl 3):294304.

  • 14

    MacDowall A, Skeppholm M, Robinson Y, Olerud C. Validation of the visual analog scale in the cervical spine. J Neurosurg Spine. 2018;28(3):227235.

  • 15

    Kim SW, Limson MA, Kim SB, Arbatin JJ, Chang KY, Park MS, et al. Comparison of radiographic changes after ACDF versus Bryan disc arthroplasty in single and bi-level cases. Eur Spine J. 2009;18(2):218231.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Miyazaki M, Hong SW, Yoon SH, Morishita Y, Wang JC. Reliability of a magnetic resonance imaging-based grading system for cervical intervertebral disc degeneration. J Spinal Disord Tech. 2008;21(4):288292.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Marques C, Granström E, MacDowall A, Moreira NC, Skeppholm M, Olerud C. Accuracy and reliability of X-ray measurements in the cervical spine. Asian Spine J. 2020;14(2):169176.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Anderson PA, Andersson GB, Arnold PM, Brodke DS, Brodt ED, Chapman JR, et al. Terminology. Spine (Phila Pa 1976).2012;37(22 suppl):S8S9.

  • 19

    MacDermid JC, Walton DM, Avery S, Blanchard A, Etruw E, McAlpine C, Goldsmith CH. Measurement properties of the neck disability index: a systematic review. J Orthop Sports Phys Ther. 2009;39(5):400417.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Pool JJ, Ostelo RW, Hoving JL, Bouter LM, de Vet HC. Minimal clinically important change of the Neck Disability Index and the Numerical Rating Scale for patients with neck pain. Spine (Phila Pa 1976).2007;32(26):30473051.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21

    White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30(4):377399.

  • 22

    R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing; 2018.Accessed October 6, 2021. https://www.r-project.org/

    • Search Google Scholar
    • Export Citation
  • 23

    Goel VK, Faizan A, Palepu V, Bhattacharya S. Parameters that effect spine biomechanics following cervical disc replacement. Eur Spine J. 2012;21(5 Suppl):S688S699.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Eck JC, Humphreys SC, Lim TH, Jeong ST, Kim JG, Hodges SD, et al. Biomechanical study on the effect of cervical spine fusion on adjacent-level intradiscal pressure and segmental motion. Spine (Phila Pa 1976).2002;27(22):24312434.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25

    Li XF, Jin LY, Liang CG, Yin HL, Song XX. Adjacent-level biomechanics after single-level anterior cervical interbody fusion with anchored zero-profile spacer versus cage-plate construct: a finite element study. BMC Surg. 2020;20(1):66.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Lou J, Li Y, Wang B, Meng Y, Gong Q, Liu H. Biomechanical evaluation of cervical disc replacement with a novel prosthesis based on the physiological curvature of endplate. J Orthop Surg Res. 2018;13(1):41.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Faizan A, Goel VK, Garfin SR, Bono CM, Serhan H, Biyani A, et al. Do design variations in the artificial disc influence cervical spine biomechanics? A finite element investigation. Eur Spine J. 2012;21(5 Suppl):S653S662.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Marques C, MacDowall A, Skeppholm M, Canto Moreira N, Olerud C. Unintended fusion in cervical artificial disk replacement: a prospective study on heterotopic ossification, progression, and clinical outcome, with 5-year follow-up. Eur Spine J. 2021;30(6):16621669.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Kong L, Ma Q, Meng F, Cao J, Yu K, Shen Y. The prevalence of heterotopic ossification among patients after cervical artificial disc replacement: a systematic review and meta-analysis. Medicine (Baltimore). 2017;96(24):e7163.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30

    Wahood W, Yolcu YU, Kerezoudis P, Goyal A, Alvi MA, Freedman BA, Bydon M. Artificial discs in cervical disc replacement: a meta-analysis for comparison of long-term outcomes. World Neurosurg. 2020;134:598613.e5.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31

    Hacker FM, Babcock RM, Hacker RJ. Very late complications of cervical arthroplasty: results of 2 controlled randomized prospective studies from a single investigator site. Spine (Phila Pa 1976).2013;38(26):22232226.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32

    Delamarter RB, Zigler J. Five-year reoperation rates, cervical total disc replacement versus fusion, results of a prospective randomized clinical trial. Spine (Phila Pa 1976).2013;38(9):711717.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33

    Sundseth J, Fredriksli OA, Kolstad F, Johnsen LG, Pripp AH, Andresen H, et al. The Norwegian Cervical Arthroplasty Trial (NORCAT): 2-year clinical outcome after single-level cervical arthroplasty versus fusion-a prospective, single-blinded, randomized, controlled multicenter study. Eur Spine J. 2017;26(4):12251235.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 2772 2772 61
Full Text Views 919 919 543
PDF Downloads 906 906 453
EPUB Downloads 0 0 0