Long-term clinical results demonstrated high clinical success rate and satisfaction rate after lumbar TDR. Kitzen et al14. reported a 79.6% satisfaction rate after lumbar TDR at a mean 12.3-years follow-up. Park et al.15 reported 76.9% clinical success rate and 87.2% satisfaction rate after lumbar TDR at mean 5 years follow-up. David et al.16 also reported 82.1% clinical success rate at 13.2 years follow-up. In an IDE study, Radcliff et al. suggested the lumbar TDR were safe and effective for single-level lumbar DDD17. They also found no significant increase of radiological presence of facet joint degeneration at 7-year follow-up17. However, many clinical studies demonstrated facet joint degeneration (FJD) after lumbar TDR, some requiring reoperation. Shim et al. observed degradation of FJD in 32% patients after ProDisc after 36–40 months and in 36.4% patients after CHARITE after 36–48 months18. Park et al. found 12 of 41 (29%) TDR levels had FJD progression after 32.2 months (26–42 months) and related with female, malposition, and 2-level TDR19. Early-results from the Norwegian Spine Study Group demonstrated that significantly more patients either had newly-onset or progressed FJD at the surgical level in the lumbar TDR group compared with the rehabilitation group (34% vs. 4%, P < 0.001) at two-year follow-up20. In a later long-term study, they found similar FJD rate at 8-year follow-up compared with the 2-year follow-up but found no association between FJD and clinical outcomes21. Pimenta et al. followed 15 patients for 7 years to find 7 patients with FJD and only 1 symptomatic22. On the contrary, some authors believed FJD was the main cause of postoperative back pain for lumbar TDR patients. Siepe et al. confirmed facet joint pain by fluoroscopically guided spine infiltrations in 9.1% patients after L4/L5 TDR, 28.1% after L5/S1 TDR, and 60% after bi-segmental TDR, using ProDisc II23. They in a later study observed that progression of FJD was present in 20% of all facet joints at 53.4 months follow-up and more common at lumbosacral junction, and progression of FJD was associated with lower ROM at the index level and inferior VAS and ODI scores24. A small portion of patients with facet joint complains eventually received revision surgery. Siepe et al. reported 29 revisions of 201 patients, among those 2 re-operations were due to facet complains25. David et al. reported 10.4% (11 of 106) reoperation rate at the index-level after lumbar TDR16. Five patients underwent revision surgery due to symptomatic FJD16. Punt et al. performed revision surgery for 75 patients after lumbar TDR, among them 25 patients (33.3%) present with FJD26. Schmitz et al. reported that 85% of the 48 patients who had revision present with FJD and concluded that FJD was the most important cause for revision after lumbar TDR27. In vitro cadaveric study and finite element analysis suggested that abnormal loading and aberrant kinematics of the facet joints after lumbar TDR contributed to the FJD28–31.
Presently, long-term results of ACDR for single-level and two-level cervical spondylosis have been published. These studies showed that ACDR, performed with different prostheses, showed higher or comparable overall successful rate and satisfaction rate, lower incidence of adjacent level degeneration, fewer revisions at either the index level or the adjacent levels, comparing to ACDF1–6. Interestingly though, these long-term studies did not specifically describe the facet joint degeneration after ACDR. Ryu et al. reported progression of FJD in 19.4% (7 of 36) levels treated with ACDR (1 of 19 Bryan and 6 of Prodisc-C) at 24 months follow-up7. They found that anterior placement of the Prodisc-C was associated with FJD. They argued that anterior placement of the Prodisc-C could increase the load on the facet joints at the index level. Meisel et al. on the other hand, in their multi-center study composed of 200 ACDR patients using Active-C, observed no FJD progression at the 4-year follow-up8. On the contrary to the limited clinical data on the post-operative alteration of the facet joints after ACDR, many cadaveric studies and finite element studies assessed the effect of ACDR on the facet joints.
Chang et al. used strain gauges provided by Vishay Micro-Measurements, Inc. in a nondestructive manner to measure the facet joint force after single-level ACDR using Prestige32. The results showed that facet joint force increased under flexion, extension, lateral bending, and axial rotation32. The results also showed 95.4% increase of the facet joint force under extension and 19.7% under flexion after insertion of the Prodisc-C prosthesis, but 6.4% decrease of and insignificant change of facet joint force under lateral bending and rotation32. Interestingly, Crawford et al. used the same strain gauge to find mild decrease of facet load during flexion and no significant change during extension33. Jaumard et al.34 and Bauman et al.35 further complicated the scenario regarding the cadaveric study of ACDR. They placed a tip pressure transducer inside the facet joint capsule without cutting it open. The results showed no significant alteration of the facet pressure in flexion and extension, but significant increase of facet pressure in ipsilateral lateral bending and torsion34,35. Partel et al. placed the film senser into the facet joint after cutting the facet capsule to directly record the facet force after C4/C5 disc replacement using Prodisc-C36. The facet forces under extension, lateral bending, and rotation were 28.75N, 55.33N, and 61.36N in the intact model, and 41.87N, 58.96N, and 58.31N in the ADCR model. The results indicated the Prodisc-C only increased the facet force under extension. These inconsistent results reflected the hardship of measuring the facet force/pressure in the cadaveric specimens. The indirect measurements by strain gauges needed complex process of calibration, yet still lacked validation. The number of gauges used, and the position of gauge placed would also significantly affect the estimation of the facet force. The film sensor could provide direct reading of the facet force but required cutting open the facet capsule. The changed biomechanics of the functional spinal unit due to cutting of capsular ligaments would therefore result in inaccurate evaluation of the effect of ACDR on facet force.
Finite element (FE) analysis provided an alternative approach to study the facet force after ACDR. Lee et al. constructed a cervical model (C2-C7) with simulated ACDR at C5/C6 level by either Prodisc-C or Mobi-C37. The results showed increased facet force about two times larger than that in the intact model under extension. The capsular ligament tension increased under flexion in the Prodisc-C model (34 MPa) and the Mobi-C model (25.8 MPa), comparing to the intact model (20 MPa). Gandhi et al. changed the material properties of the IVD to simulate its degenerative state and studied the post-operative biomechanics of the cervical spine after ACDR by either Bryan or Prestige LP at C5/C6 level38. The results demonstrated considerable increase of facet force under all loadings except left lateral bending in the Prestige LP model. A team from Medical College of Wisconsin presented a series of FE studies comparing the biomechanical effect of different prostheses, including Bryan, Prodisc-C, Prestige LP, Mobi-C and Secure-C39–42. The facet force increased in all ACDR models under extension, with Bryan and Secure-C to a lesser extend while Mobi-C, Prestige LP and Prodisc-C to a greater extend. Besides, comparing to the intact model, facet force increased under flexion for Prodisc-C, Prestige LP, increased under lateral bending for Bryan, Prestige LP, Mobi-C and Secure-C, all to various degrees.
These prostheses differed in structure design, number of components, bearing surfaces, and articulation design. Kinematic degrees of freedom, bult-in stiffness, and patients- or surgical-related factors all played crucial parts in the postoperative biomechanics of the cervical spine43. Presently, center of rotation, one of the prosthetic traits, has been meticulously studied. Ahn et al. simulated three types of CORs and studied their impact on the cervical facet force, a fixed COR at the disc level, a fixed COR below the endplate (6.5 mm below the disc level), and a mobile COR at the level44. The results showed that the fixed COR at the disc level considerably increased the facet force under extension and lateral bending (364.5 N and 104.9 N) comparing to the intact model (14.3 N and 51.5 N)44. The lower fixed COR increased facet force under extension to a lesser degree (91 N), whereas the mobile COR did not build up the facet force under all loadings44. Galbusera et al. showed that when the COR was fixed, lower the COR (close to the physiological position under the endplate) would result in lower facet force in extension and lateral bending45. Rousseau et al. corroborated this finding by showing the posterior and lower positioned COR result in smallest facet force46. Faizan et al. showed that the facet forces were smaller for the design with inferior ball component indicating lower position of the COR47. Mo et al. showed that prosthesis with mobile COR resulted in smaller maximum facet stress compared to the fixed COR48. In short, these studies suggested that ACDR would result in increased facet force at the index level, which was believed to be a risk factor for development or progression of FJD. The increased FCFs could cause micro-injury to the facet joints. The accumulation of these micro-injuries with daily neck activity in time could therefore initiate or accelerate the FJD process.
Apart from the prosthesis- or surgical-related factors, here we present an anatomical-related factor that could alter the facet force and facet capsule strains after ACDR. We previously reported that facet tropism could cause increase of FCFs comparing to symmetrical model10. This finding was in accordance with previous biomechanical and FE studies49–51. Further, in this study we showed that, ACDR together with facet tropism, could magnify the abnormal facet loading caused by facet tropism alone. In theory, abnormally increased facet loadings could be related to the FJD. Though evidence regarding the effect of facet tropism on FJD after ACDR was not available now, data from the lumbar spine supported the hypothesis that facet tropism could be associated with FJD progression after total disc replacement (TDR). Shin et al. observed that the FJD progression levels had significantly larger facet tropism than the non-FJD progression levels at the 36-month follow-up after lumbar TDR using ProDisc-L52. Besides, we previously observed higher facet joint degeneration rate at the cervical level with facet tropism in cervical spondylosis patients9. Yet, long-term observation on the alteration of the facet joints after ACDR are needed to validate such theory in the cervical spine.
This study has some limitations. First, the cartilages of the opposing facet joints were simulated as flat components. In the real world, the surface of facet joints has various shapes. The flat type was only one of many. However, the purpose of this study was to examine the effect of facet tropism on facet stress distribution, and the flat form of the facet joint was the most straightforward candidate for presentation. In certain circumstances, a more complex analysis may be necessary. Second, the cervical model was based on a young male with no symptoms. The results should be applied with caution due to the possibility that degenerative changes of the cervical spine would complicate their interpretation. Thirdly, only the Prestige LP cervical disc replacement was evaluated in this study. Prestige LP was considered a semi-constrained artificial disc. Constrained versus unconstrained artificial discs, such as Prodisc-C vivo or Mobi-C, might have a distinct effect on the cervical spine in the presence of facet tropism. Additional investigation is necessary to determine the influence of artificial disc design on facet stress distribution when facet tropism is present.
The existence of face tropism could considerably increase facet contact force and facet capsule strain after ACDR, especially under extension, lateral bending, and rotation. Facet tropism also could result in abnormal stress distribution on the facet joint surface and facet joint capsule. Such abnormality might be a risk factor for post-operative facet joint degeneration progression after ACDR, which needs long-term clinical study to verify. Nevertheless, facet tropism might worth paying attention to when ACDR was considered as the surgical option.