Biomechanical Evaluation of Anterior Plate Fixation With Cage for Basilar Invagination With Atlantoaxial Dislocation: A Cadaveric Study
Article information
Abstract
Objective
To evaluate the biomechanical characteristics of 2 anterior fixation techniques (clival plate fixation [CPF], transoral atlantoaxial reduction plate [TARP]) versus posterior occipitocervical fixation (POCF) for basilar invagination with atlantoaxial dislocation (BI-AAD), under varying atlantoaxial lateral mass cage heights (4–10 mm).
Methods
Seven fresh cadaveric specimens (occiput to C3, Oc–C3) were tested in the following conditions: (1) intact state; (2) BI-AAD state; (3) BI-AAD+CPF; (4) BI-AAD+TARP fixation; (5) BI-AAD+POCF. A pure 1.5 N·m moment loads to specimens in flexion/extension, lateral bending and axial rotation. Range of motion (ROM) and neutral zone (NZ) values at Oc–C2 were calculated and compared.
Results
ROM of the C1–2 segment under the intact and BI-AAD states were as follows: 9.3°±4.6° versus 21.3°±8.3° in flexion, 4.6°±1.9° versus 9.3°±3.8° in extension, 3.6°±2.2° versus 12.0°±6.5° in lateral bending, and 68.9°±14.4° versus 76.6°±6.6° in axial rotation, respectively. Compared with BI-AAD states, all internal fixation techniques significantly reduced the ROM of the Oc–C2 segment. TARP fixation exhibited larger ROM in flexion-extension. While in lateral bending and axial rotation, the ROM values for the anterior plate constructs were smaller than that of POCF, with a statistically significant difference observed between CPF and POCF. Cage height variations showed no significant impact on overall biomechanical stability.
Conclusion
Anterior plate fixation techniques demonstrated superior resistance to lateral bending and rotational forces compared to posterior approaches, with clival plate fixation exhibiting optimal biomechanical stability for BI-AAD. Variations in cage height exhibited negligible impact on stability when internal fixation achieved adequate rigidity.
INTRODUCTION
Basilar invagination combined with atlantoaxial dislocation (BI-AAD) represents a complex craniocervical junction deformity characterized by progressive dorsocranial migration of the odontoid process, which can lead to severe compression of the brainstem and spinal cord, resulting in significant neurological deficits such as myelopathy, motor weakness, and sensory disturbances and requiring surgical management [1-3]. Epidemiologically, this condition frequently coexists with congenital atlas occipitalization (AOZ), with AOZ prevalence rates of 50%–92% in BI-AAD populations compared to 0.084%–3.630% in the general populace [2,4-6]. Our prior investigations have identified critical anatomical challenges in these patients, including osseous hypoplasia of the occipital squama, reduced C2 pedicle dimensions, and high prevalence of vertebral artery anomalies—factors complicating posterior instrumentation strategies [7,8].
Currently, surgical approaches for craniocervical junction pathologies include anterior transoral, posterior cervical, and combined techniques. Although combined approaches can achieve effective neural decompression [9], they require intraoperative repositioning, which increases surgical trauma, prolongs operative time, and elevates the risk of spinal cord injury. Posterior reduction and instrumented fusion is the most widely adopted strategy. Advocates of this strategy emphasize its simplicity and ease of execution, as well as its satisfactory reduction and realignment capabilities when combined with modern surgical techniques and instrumentation [10-13]. However, this approach is not suitable for patients with posterior structural deficiencies or those with BI-AAD and coexisting AOZ. Furthermore, conventional posterior approaches may fail to adequately address ventral spinal cord compression without direct decompression. The transoral atlantoaxial reduction plate (TARP) system addresses ventral compression through direct decompression and anatomical reduction [14,15], but faces limitations in occipitocervical stabilization as it does not inherently stabilize the occipitoatlantal (C0–1) joint, which is crucial in cases with BI-AAD and AOZ. Building upon these limitations and our previous research findings [16], we propose the clival plate fixation (CPF) as a potential alternative approach for the reduction and fixation of BIAAD and AOZ. However, the comparative efficacy of these fixation techniques in patients with BI-AAD and AOZ remains unclear and warrants further investigation.
Fusion cages have been widely utilized in cervical and lumbar spinal fusion procedures. In 2007, Goel [17] introduced a pioneering technique involving the insertion of a titanium spacer into the distracted lateral mass joint of the atlantoaxial complex to achieve joint distraction and realignment. Clinically, this principle has been foundational to facilitate reduction and fusion of the atlantoaxial joint, as well as caudal repositioning of the odontoid process, thereby alleviating neural compression.
Given the limited direct comparisons of these internal fixation techniques (CPF, TARP fixation, posterior occipitocervical fixation [POCF]) within a single study and the lack of biomechanical evidence regarding the influence of cage height on construct stability, we conducted the present cadaveric study. The objective was to quantitatively evaluate the biomechanical differences among 2 anterior plate fixation techniques versus POCF for treating BI-AAD with AOZ under varying cage heights. This preclinical biomechanical investigation also serves as an essential step to validate the feasibility and stability of a novel technique before it can be considered for clinical application.
MATERIALS AND METHODS
1. Specimen Preparation and Ethics Statement
This study was approved by the Ethic Committee of Nanfang Hospital (NFEC-2022-197). The study was conducted in accordance with the principles of the Declaration of Helsinki.
Seven fresh human cadaveric spinal specimens (Oc–C3) were obtained, with average age of 45 years (range, 30–65 years; 1 unknown age; 5 males and 2 females). All specimens underwent a general examination and radiograph images to exclude osseous malformations, structural anomalies, or defects in the craniocervical junction, and specimens with osteoporosis (T value<2.5) were excluded by bone mineral density examination. Fresh specimens were sealed in dual-layer plastic bags and cryopreserved at -20°C. Before the experiment, specimens were thawed to ambient temperature (25°C). The muscle tissue was carefully removed while the joint capsules and ligaments were carefully preserved. The C3 vertebrae and partial occiput were immobilized in dental plaster, maintaining the articular surfaces of atlantoaxial lateral mass joint in a neutral horizontal plane.
2. Construction of BI-AAD With AOZ In Vitro Model
The construction of the specimen model was performed following intact state testing and involved the following procedures: (1) Atlanto-occipital joint fixation: bilateral atlanto-occipital articular surfaces were uniformly resected using a 5-mm grinding burr, with 2.5 mm removed from both superior and inferior articular surfaces along the joint space orientation to create a 5-mm vertical interarticular gap. Posterior screw fixation of the atlanto-occipital joint was implemented in 20° flexion to simulate fused status. (2) Transection of the transverse atlantal ligament to facilitate mild anterior tilt dislocation of the atlantoaxial joints. Radiographic analysis confirmed successful modeling through quantitative criteria: odontoid tip distance to the McRae line (foramen magnum anterior-posterior margin connection) <5.8 mm and atlantodental interval >3 mm. Model validation was achieved upon fulfillment of these radiographic parameters (Fig. 1B). A representative photograph of the constructed specimen model is shown in Fig. 1A.
(A) BI-AAD with AOZ model in vitro: anterior and lateral views. (B) X-ray of BI-AAD with AOZ model in vitro. BIAAD, basilar invagination with atlantoaxial dislocation; AOZ, atlas occipitalization.
3. Design of Plates and Cages
The clival plate was independently designed by our research team and has obtained a patent application (Patent authorization number: ZL 201410821600.4). The material of the anterior plates was stainless steel. The atlantoaxial lateral mass fusion cage was developed based on computed tomography anatomical measurements and clinical application studies [18]. To optimize implant compatibility, the cage adopted a rectangular prism configuration, with a bullet-shaped streamlined profile at its anterior end to minimize insertion resistance. For enhanced biomechanical stability, a serrated antislip structure was integrated into the midsection of the device, which increases the frictional coefficient at the bone-implant interface to resist displacement. The cage was made of photopolymer and dimensions were 11 mm in length, 8 mm in width, and a graded height design (4, 6, 8, 10 mm) (Fig. 2B). During radiographic imaging, a 0.8-mm diameter iron wire was fixed along the central longitudinal axis of the cage to facilitate visualization of its positioning. Internal fixation devices and surgical operation instruments were all provided by Guangzhou Dekang Medical Device Technology Co., Ltd. Plates and fusion cages used in the experiment were illustrated in Fig. 2.
The required instruments during the experiment: anterior plates (A) and cages (B).
4. Surgical Techniques
During anterior CPF, a clival plate matching the anatomical contour of the clival bone surface was selected to ensure tight contact between its clivus portion and the osseous surface, with the caudal end extending to the midinferior border of the C2 vertebral body. A 1.0-mm diameter Kirschner wire was used to predrill pilot holes perpendicular to the bone surface through the screw holes on the clivus portion of the plate, penetrating the contralateral cortical margin. Subsequently, a 2.5-mm diameter drill bit was employed to create threaded holes, followed by reaming with a 3.5-mm diameter tap. Bilateral atlantoaxial lateral mass fusion cages with appropriate heights were implanted according to experimental requirements. Following measurement of screw trajectory lengths (10–15 mm), 3 screws were inserted into the clival fixation segment. For atlas lateral mass screw placement, the screw holes on the atlas segment of the plate served as the entry points, with a 10° lateral inclination trajectory adopted for bilateral screw insertion. Axis fixation utilized a 4-screw technique: 2 subarticular screws were positioned immediately inferior to the C2 superior articular facets in the horizontal plane, and 2 transverse body screws were inserted bilaterally along the midline of the C2 vertebral body in the horizontal orientation (Fig. 3A). Postfixation radiographic verification confirmed proper implant positioning (Fig. 4A).
Flexibility test of the upper cervical spine specimen implanted with internal fixation devices. (A) Clival plate fixation state. (B) Transoral atlantoaxial reduction plate fixation state. (C) Posterior occipitocervical fixation state. The specimen was mounted on the custom-designed spine testing machine with the C3 vertebrae secured to the base of the machine. A pure moment of 1.5 Nm was applied to the specimen in flexion-extension. Three-dimensional motion of the occiput and vertebrae was recorded via marker carriers of an optoelectric camera system. 1, occipital bone; 2, atlas; 3, axis.
Radiographic anterior-posterior and lateral views of 3 fixation devices with cages of 10 mm in height implanted in the atlantoaxial lateral mass joints. (A) Clival plate fixation. (B) Transoral atlantoaxial reduction plate fixation. (C) Posterior occipitocervical fixation. White arrows indicate the position of the fusion cages.
The TARP fixation procedure was performed following similar technical principles to the atlas-axis operative steps described for CPF. Plate and fusion cage selection was performed according to experimental requirements to achieve optimal fixation (Fig. 3B). Postoperative radiographic verification confirmed appropriate implant alignment (Fig. 4B).
During POCF (Peak Summit, Depuy Spine), 3 occipital screws were inserted after midline drilling and tapping on the occipital bone to secure the occipital plate. C2 fixation employed a pedicle screw technique, with the entry point positioned 5 mm inferior to the superior border of the lamina and 7 mm lateral to the medial margin of the spinal canal. Screw trajectories were oriented at 20° cephalad angulation in the sagittal plane and 30° medial inclination in the coronal plane. Bilateral atlantoaxial lateral mass fusion cages of appropriate heights were implanted based on experimental specifications. Following screw placement, a 3.5-mm diameter titanium rod was contoured to match the physiological curvature and sequentially connected to the occipital plate and C2 pedicle screws to achieve rigid occipitocervical stabilization (Fig. 3C). Postoperative radiographic evaluation confirmed proper anatomical alignment of the implants (Fig. 4C).
5. Flexibility Test
This study utilized a repeated-measures design with each specimen serving as its own control, tested sequentially under the following conditions: (1) intact state; (2) BI-AAD state; (3) BI-AAD+CPF; (4) BI-AAD+TARP fixation; (5) BI-AAD+POCF. To mitigate potential screw-bone interface loosening caused by repeated plate installation/removal during testing, screw trajectories were reinforced with ultrahard dental base resin during the second fixation cycle. Crucially, the testing order for 3 fixation conditions was randomized to counteract potential sequence effects, with fusion cages (4/6/8/10 mm in height) randomly assigned during motion testing. While a sample size of 7 specimens may seem limited, it is consistent with standards in cadaveric spine biomechanics. The use of a repeated-measures design, where each specimen serves as its own control, significantly increases the statistical power by minimizing interspecimen variability. The sequence of flexion-extension, lateral bending, and axial rotation testing was similarly randomized. All tests for a single specimen were completed within 24 hours.
A custom-designed spine testing apparatus applied pure 1.5 N·m moment loads to specimens in flexion/extension, left/right lateral bending, and left/right axial rotation. The servo motor-driven system delivered 3 continuous loading cycles at 2°/sec for each motion. Ambient temperature was maintained at 25°C throughout testing, with specimens periodically irrigated with physiological saline to preserve hydration.
Motion measurement was performed using the Optotrak Certus 3D motion measurement system (Northern Digital Inc., Canada). This high-precision automated system undergoes a manufacturer-specified calibration prior to each testing session to ensure accuracy (documented accuracy <0.1 mm), thereby minimizing measurement error and making inter- or intraobserver variability negligible. Four 2.0-mm diameter Kirschner wires were surgically implanted into the occipital bone and posterior/lateral aspects of C1–3 vertebrae, with strict intraoperative control of insertion depth and trajectory alignment to avoid interference with pedicle screw pathways. Each Kirschner wire was attached to a customized infrared marker array containing 4 active markers (Fig. 3). Motion data were acquired at 20-Hz sampling frequency through continuous tracking of marker trajectories. The following biomechanical parameters were derived from system analysis: neutral zone (NZ), defined as the motion range during recovery from unloaded state to neutral position; and range of motion (ROM), representing the total displacement from neutral position to maximum loading state. Specifically, ROM and NZ values at Oc–C2 segment were analyzed.
6. Statistical Analysis
Based on the biomechanical symmetry of lateral bending and rotational motion in the left-right directions, the kinematic parameters of the same side were averaged for statistical analysis. Data processing was conducted using IBM SPSS Statistics ver. 27.0 (IBM Co., USA). Repeated-measures analysis of variance was employed for comparisons among multiple groups, and the Student-Newman-Keuls (SNK) post hoc test was applied for pairwise comparisons between groups. The assumptions of normality for the data were verified prior to analysis. While the SNK test is a powerful tool for exploratory analysis, we acknowledge its higher propensity for type I errors compared to more conservative tests; therefore, the results should be interpreted with this consideration. ROM and NZ values were expressed as mean±standard deviation. The significance level was set at p<0.05.
RESULTS
No loosening or mismatch was observed during testing. ROM and NZ values of the Oc–C2 segment in different states were graphically displayed in Figs. 5 and 6, and specific data were shown in Tables 1 and 2. Specific p-values are reported in Table 3.
Range of motion (ROM) comparison of Oc–C2 segment during different rotation under 3 distinct fixation techniques following cage implantation with varying heights at the atlantoaxial joint (*p<0.05, **p<0.01; Specific p-values are reported in Table 3). CPF, clival plate fixation; TARP, transoral atlantoaxial reduction plate; POCF, posterior occipitocervical fixation with occipital plate and C2 pedicle screws. Data are presented as mean and standard deviation (n=7). Statistical comparisons were made using repeated-measures analysis of variance with a Student-Newman-Keuls post hoc test.
Neutral zone (NZ) comparison of Oc–C2 segment during different rotation under 3 distinct fixation techniques following cage implantation with varying heights at the atlantoaxial joint (*p<0.05, **p<0.01; Specific p-values are reported in Table 3). CPF, clival plate fixation; TARP, transoral atlantoaxial reduction plate; POCF, posterior occipitocervical fixation with occipital plate and C2 pedicle screws. Data are presented as mean and standard deviation (n=7). Statistical comparisons were made using repeated-measures analysis of variance with a Student-Newman-Keuls post hoc test.
ROM and NZ values in the Oc–C2 and C1–2 (n=7)
ROM and NZ values between the occiput and C2 (n=7)
Statistical comparison p-values from repeated-measures ANOVA with SNK post hoc tests (n=7)
In this study, ROM of the C1–2 segment under the intact and BI-AAD states were as follows: 9.3°±4.6° versus 21.3°±8.3° in flexion, 4.6°±1.9° versus 9.3°±3.8° in extension, 3.6°±2.2° versus 12.0°±6.5° in lateral bending, and 68.9°±14.4° versus 76.6°±6.6° in axial rotation, respectively. Compared to the intact state, the BI-AAD state demonstrated significantly increased ROM in flexion, extension, and lateral bending, all showing statistically significant differences. Although a slight increase in ROM was also observed in axial rotation, the difference was not statistically significant (Table 1). Due to the fusion of the occipitoatlantal joint during the construction of the BI-AAD model, the ROM of the C1–2 segment in the BI-AAD state represents that of the Oc–C2 segment. Compared with the intact and BI-AAD state, all 3 internal fixation techniques significantly reduced the ROM of the Oc–C2 segment. In flexion, the TARP fixation exhibited a relatively larger ROM, while the CPF and POCF showed comparable ROM values, both lower than that of TARP fixation. In extension, there was no statistically significant difference in ROM among the 3 fixation constructs; however, CPF showed the smallest ROM. During lateral bending and axial rotation, the ROM values of both CPF and TARP fixation were smaller than that of POCF, with a statistically significant difference observed between CPF and POCF (Fig. 5). Notably, no significant differences in overall biomechanical stability were observed among the constructs with different cage heights implanted at the atlantoaxial joint (Table 2).
Compared with the intact state, the BI-AAD state showed an increase in the NZ of the C1–2 segment in flexion-extension and lateral bending, while a slight decrease was observed in axial rotation. However, none of these changes reached statistical significance (Table 1). Following the implantation of the 3 fixation constructs, the NZ values of the Oc–C2 segment were significantly reduced in all motion directions compared to the BI-AAD state. Among the 3 fixation techniques, TARP fixation exhibited a significantly larger NZ in the flexion-extension direction compared to the other 2 techniques. In contrast, no statistically significant differences in NZ were observed among the 3 fixation techniques in lateral bending or axial rotation (Table 2, Fig. 6).
DISCUSSION
The etiological factors underlying BI-AAD are multifactorial and complex, primarily encompassing congenital developmental anomalies (occipital basilar hypoplasia, AOZ, and odontoid process malformations), acquired contributors (including trauma, inflammatory conditions, metabolic disorders, and neoplasms), degenerative alterations (such as transverse ligament degeneration and atlantoaxial joint degeneration), and iatrogenic causes (surgical complications and postradiotherapy structural changes) [19,20]. These etiological factors collectively disrupt the osseous architecture and/or ligamentous stability at the craniocervical junction, ultimately culminating in BI-AAD. The pathomechanical destruction can be categorized into 2 principal mechanisms: (1) Osseous structural compromise, including fractures of the occipital condyles, atlas/axis vertebral arches, and odontoid process fractures. Notably, odontoid fractures serve as a pivotal destabilizing factor in atlantoaxial dislocation pathogenesis. (2) Ligamentous insufficiency, predominantly characterized by transverse ligament rupture. Concurrent osseous and ligamentous injuries synergistically exacerbate craniocervical instability. Moreover, AOZ was widely documented as the most prevalent craniocervical junction malformation [21,22], representing a pivotal etiological contributor to BI-AAD. Mechanistic studies by Guan et al. [23] revealed that AOZ induced a reduction in C1 lateral mass height leading to odontoid process displacement, thereby mediating upward migration characteristic of basilar invagination. Furthermore, emerging evidence suggests that AOZ may destabilize the craniocervical architecture through biomechanical cascades: degenerative remodeling of the lateral mass joints and subsequent spondylolisthesis have been implicated in the progressive AAD [24]. These pathoanatomical alterations collectively compromise the osseoligamentous integrity of the craniovertebral junction, establishing AOZ as a critical nexus in BI-AAD pathophysiology. To recapitulate the clinical instability profile of BI-AAD, this study developed a composite injury model integrating both osseous destruction and ligamentous disruption. The model was established through resection of the atlanto-occipital joint capsule and anterior atlantal membrane, reduction of atlanto-occipital joint surface height via bone grinding, arthrodesis under anteriorly flexed head positioning, and transection of the transverse ligament. This methodology effectively simulates the biomechanical instability patterns observed in clinical BI-AAD presentations. Previous study has employed ligament resection techniques to establish experimental models of atlantoaxial instability [18]. However, biomechanical investigations typically utilize specimens derived from nonpathological donors, necessitating model development in anatomically intact substrates. This methodological paradigm introduces inherent limitations, as the artificially induced instability fails to fully replicate the complex pathoanatomical alterations observed in clinical disease states. Consequently, model validation remains constrained to radiographic parameter quantification rather than comprehensive pathophysiological emulation.
Goel proposed that instability constituted the primary etiological determinant of BI, advocating stabilization as the cornerstone therapeutic intervention [25]. Current surgical approaches encompass anterior transoral, posterior cervical, and combined techniques. Posterior reduction-fixation procedures are widely favored due to their technical simplicity, operational feasibility, and favorable clinical outcomes. Yin et al. [12] documented 174 cases of irreducible AAD with BI treated via posterior reduction-fixation, achieving complete or >90% reduction in 62.9% of patients, 60%–90% reduction in 30%, and <50% reduction in 7.1%, with significant postoperative neurological improvement. However, conventional posterior approaches exhibit inherent limitations in addressing craniocervical pathologies, particularly in cases of ventral spinal cord compression, where they fail to achieve adequate decompression, anatomical realignment, or restoration of physiological biomechanics [26]. Consequently, combined anterior-posterior approaches have emerged as the clinical gold standard, enabling staged or simultaneous anterior decompression and posterior stabilization. Srivastava et al. [9] reported 19 cases of irreducible AAD with BI treated via anterior release and posterior reduction-fixation, achieving anatomical reduction in 78.9% and partial reduction in 21.1%, with significant improvements in cervicomedullary angle and Japanese Orthopaedic Association scores. Despite these advancements, combined approaches necessitate intraoperative positional changes, prolong surgical duration, increase trauma, and elevate risks of iatrogenic spinal cord injury during patient repositioning [13]. The evolution of transoral techniques has revolutionized ventral craniocervical access, with the TARP system demonstrating superior clinical efficacy in ventral decompression [27-30]. Comparative studies confirm that TARP and occipitocervical fusion effectively address BI with irreducible AAD, though TARP outperforms occipitocervical fusion in reduction quality, decompression adequacy, and early fusion rates [31]. Critics note limitations in standalone anterior approaches, particularly regarding implant stability: the complex pathoanatomy of BI-AAD, compounded by osteoporosis, challenges secure anterior plate-screw fixation [32]. Furthermore, isolated atlantoaxial stabilization proves insufficient for cases with concurrent occipitoatlantal instability. Building on anatomical investigations revealing thickened clival bone in BI-AAD patients—a potential anterior fixation anchor [16], we propose CPF incorporating clival plate and atlanto-occipital joint screws to enhance biomechanical stability [33]. Nevertheless, comparative biomechanical analyses of these fixation techniques remain sparse. A finite element study suggested TARP fixation may confer superior stability in extension, lateral bending, and axial rotation compared to posterior C1–2 rod-screw constructs, albeit with reduced flexion resistance [34]. For the first time, we compared the stability of 3 internal fixation techniques (CPF, TARP fixation, and POCF) in the same biomechanical study. The results of the study showed that all 3 fixation techniques effectively reduced ROM beyond that of both the intact and BI-AAD states, indicating satisfactory immediate postoperative stability. However, the long-term fusion outcomes require further investigation. Biomechanical testing demonstrated that, during flexion, the TARP fixation exhibited a relatively larger ROM, whereas CPF and POCF exhibited comparable and significantly lower ROM values. This is consistent with the results of the previously mentioned finite element study. In extension, no statistically significant differences in ROM were observed among the 3 fixation techniques; however, CPF showed the smallest ROM. In flexion-extension overall, CPF consistently demonstrated a smaller ROM compared to both TARP and posterior fixation, suggesting superior stabilization while TARP fixation showed poor stability among the 3 techniques. The enhanced performance of CPF may be attributed to the unique structural design of the anterior clival plate, particularly its clival geometry and multipoint fixation strategy. The relatively greater ROM observed with TARP fixation in flexion may result from its limited contact area and fewer fixation points. In lateral bending and axial rotation, CPF again demonstrated the smallest ROM, with statistically significant differences observed between CPF and POCF. Among the 3 internal fixation techniques, posterior fixation showed poor resistance to lateral bending as well as to rotation. Although the posterior occipitocervical construct involves a greater number of fixation points, the mechanical load distribution may be less optimal, limiting its ability to effectively constrain motion. These findings suggested that CPF provided superior multidirectional motion control, effectively restricting abnormal spinal movement and offering enhanced biomechanical stability. Consequently, it may create a more favorable mechanical environment for achieving successful osseous fusion and be used as a supplementary technique for anterior occipitocervical fixation.
The implantation of atlantoaxial lateral mass fusion cages has emerged as a prevalent surgical strategy for achieving reduction and fusion in BI-AAD. Biomechanical studies demonstrate that fusion cages not only stabilize the atlantoaxial joint but also compensate for vertical height loss secondary to occipitocervical fusion [35]. Postimplantation odontoid descent facilitates neural decompression, with clinical evidence confirming its therapeutic efficacy in BI resolution [36]. Notably, the magnitude of odontoid descent correlates significantly with surgical outcomes [37]. However, the biomechanical implications of cage implantation on fixation stability remain understudied. Contradicting conventional assumptions, Li et al. [18] revealed through comparative biomechanical testing that cage implantation failed to enhance stability compared to standalone C1–2 pedicle screw fixation. Addressing this knowledge gap, our study pioneers the biomechanical evaluation of 3 fixation constructs under varying cage heights (4 mm, 6 mm, 8 mm, and 10 mm). Results demonstrated no significant differences in biomechanical stability across cage heights for all constructs, which may attribute to 2 key factors: (1) Dominance of fixation rigidity: All systems exhibited sufficient mechanical resistance to atlantoaxial motion, whether through the triple clival screws fixation, bilateral transarticular screws of TARP, or posterior constructs with occipital plates and dual-axis screws. (2) Height-dependent local effects: Cage height variations predominantly influenced bone graft contact area and regional stress distribution rather than global construct stability. These findings suggest that cage height selection should prioritize anatomical compatibility and reduction requirements rather than stability optimization when robust internal fixation is achieved. Clinically, this paradigm shift permits tailored cage sizing based on individual patient morphology. As an in vitro investigation, this study did not assess the impact of cage height on long-term fusion rates—a critical clinical endpoint requiring validation through prospective cohort studies. Furthermore, the biomechanical model excluded dynamic factors such as cyclic loading and osseointegration processes.
In this study, the measured ROM of the Oc–C2 segment under the intact state closely aligns with the physiological mobility reported by Nassos et al. [38] Comparative biomechanical evaluation revealed significant alterations in ROM of C1–2 segment between intact state and unstable BI-AAD model state. The unstable state demonstrated statistically significant increases ROM of C1–2 segment in flexion, extension, and lateral bending compared to the intact state (p<0.05). While ROM in axial rotation exhibited a nominal increase in the unstable group, this difference did not reach statistical significance. Previous studies have also reported that patients with craniovertebral junction deformities often present with clinically restricted cervical rotational motion [39]. This phenomenon may be attributed to BI-AAD combined frequently with severe AOZ, with the underlying pathophysiology primarily involving joint subluxation, lesions of muscle ligaments and subsequent limitation of cervical mobility—findings that are consistent with the motion characteristics observed in the BI-AAD model.
This study has several limitations. First, while our self-controlled design minimized interspecimen variability, the experimental design did not account for biological variables such as donor age and sex, which can influence fixation strength. Second, our biomechanical model, while necessary for standardized comparison, is an artificial simplification of a complex congenital condition and does not fully replicate the chronic pathoanatomical changes of clinical BI-AAD with AOZ. Third, the sequential testing protocol on each specimen, despite mitigation efforts like randomization and screw tract reinforcement with dental resin, may introduce confounding effects from cumulative microdamage. However, an ideal experimental protocol would limit the number of procedures per specimen, which would, in turn, significantly increase the number of required specimens. Fourth, this study only assessed immediate postoperative stability. It did not include cyclic or fatigue loading, which is critical for evaluating the long-term durability of the constructs and their resistance to failure under repetitive physiological stress. Long-term fusion outcomes and biological responses can only be evaluated through future in vivo studies. Fifth, an a priori power analysis was not conducted to determine the sample size, a common issue in cadaveric research due to specimen scarcity. However, the study successfully detected significant differences, suggesting the power was adequate for the primary outcomes. Finally, the anterior plate used was a standardized design. Given the significant anatomical diversity in patients with BI-AAD and AOZ, clinical application would likely require modular or patient-specific implants to ensure proper fit and function. This study validates the biomechanical principle, but implant design remains a topic for future research.
CONCLUSION
Compared with POCF, anterior plate fixation demonstrated greater resistance to lateral bending and rotation. Variations in cage height exhibited no significant impact on biomechanical stability when internal fixation achieved adequate rigidity. CPF, as a potential alternative technique, demonstrated superior biomechanical performance in BI-AAD stabilization compared to TARP and posterior occipitocervica fixation, providing a promising biomechanical rationale for further investigation.
Notes
Conflict of Interest
The authors have nothing to disclose.
Funding
This research was supported by the National Natural Science Foundation of China (No. 82172523), Natural Science Foundation of Guangdong Province (2022A1515010488).
Author Contribution
Conceptualization: WJ; Formal Analysis: JZ, ZH; Investigation: JZ, ZH, KL, XZ, JX, YW, JX, PX; Methodology: JZ, WJ, ZH; Project Administration: JZ, ZZ, WJ; Writing – Original Draft: JZ; Writing – Review & Editing: WJ, JZ.