Objective To study biomechanic characteristics of the atlantoaxial joint.
Methods The bone-ligament specimens from 6 fresh cadaveric occiput-C3 were used and the ranges of three-demensional rotation movements at the C1-C2 joint was quantitatively determined by using a maximum of 1.53 Nm.
Results Neutral zones for flexion/extention, left/right lateral bending, and left/right axial rotation were 15.00,20.29, 7.25, and 47.93 degrees, respectively. The greatest intervertebral motion was axial rotation at the C1-C2 joint, with the neutral zone constituting 75.99%. The ranges of coupled axial rotational motion of flexion, extension, left lateral bending, and right lateral bending were 12.42, 16.14, 22.07, and 22.48 degrees, respectively.
Conclusion Congenital anomalies (atlanto-occipital fusion, dentes deformity, congenital loss of transverse ligament of atlas besides trauma and infection (pharyngealthroatitis, rheumatism) have a risk of displacement of C1 and C2. When the C1-C2 joint is fused for the treatment of atlantoaxial anterior dislocation, both anterior gliding of the atlas and overcoming rotation instability of the C1-C2 joint are prevented, therefore, bony fusion is gained between bone blocks and C1-C2.

Trauma and disease of the atlantoaxial joint are commonly seen clinically. Clinical studies on the operational methods of atlas dislocation are available in literature. However, these methods were not based on biomechanical experiments but on unverified hypothesis. In the present study, the bone-ligament specimens from 6 fresh cadaveric C0-C3 were obtained using a maximum of 1.53 Nm, the ranges of three-dimensional physiologic rotation movements at the atlantoaxial joint were quantitatively determined and the clinical importance is discussed.

MATERIALS AND METHODS
Specimen Preparation ix fresh human cadaveric occipitoatlantoaxial complex specimens were harvested from 6 young men, who died of acute craniocerebral injury, with average age of 21 years (ranging 19-28 years). Specimens were radiographed, which showed no bony abnormalities or degenerative changes. Each specimen, consisting of the base of the occiput (CO), C1, C2 and C3 vertebrae, was cleared of all the muscular tissue, and the entire ligment was carefully preserved. C0 and C3 were embedded respectively in epoxy block with jig. The posterior wall of the C3 vertebral body was inclined anteriorly by 20 degrees and C0 was made horizontal.¹ The anterior walls of the C0, C1, and C3 vertebral body were inlayed separately by the three specifically designed studs, one aspect exposing as marker points. These vertebral markers were designed so that throughout the range of motion no marker contacted each other and were always visible.

RESULTS
Ranges of the main motion, neutral zones and elastic zones for the atlantoaxial joint are given in Table 1. In lateral bending the average one-side rotation was 7.25 degrees. In the sagittal plane, the motion included flexion (15.00 degrees) and extension (20.29 degrees). The maximum motion in this joint was in axial rotation, averaging 47.93 degrees to either side, with very little asymmetry. It was noticed that the extremely large neutral zone constituted 75.99% of this ranges of motion.

Table 1 C1-C2 joint main motion (degree, ±s)

Table 2 C1-C2 joint coupled motion (degree, ±s)

The upper and lower parts of specimen were embedded in quick-setting epoxy cast; the entire C0-C3 were used to replace the single atlantoaxial joint. The experiment imitated the physiological way of motion of the occipitatlantoaxial complex: flexion, extension, lateral bending and axial rotation.

The spinal behavior is highly nonlinear and viscoelastic.¹ By allowing 30 second creep, the viscous effects were minimized. The nonlinearity was documented by dividing the range of motion into two parts: neutral zone, a measure of initial low stiffness behavior; elastic zone, representing the resistant phase of the elastic behavior. Three motion parameters were defined in the third load cycle: neutral zone-displacement measured from the neutral position to the zero load-step, elastic zone-displacement measured from the zero load-step to the maximum load-step and range of motion-sum of the neutral and elastic zones. Because of the difficulty of precisely defining the neutral position, this point was assumed to lie midway between the zero-load points of the positive and negative moments of a moment pair.
The present study confirmed that the special design of the articulations and ligaments together provide the specific movements of this region. Dickman et al.⁴ thought the transverse ligament is the largest, strongest, and thickest ligament, but nonelastic and rather rigid at the C0-C2 joint complex. The transverse ligament is the strongest ligament stabilizing the atlas, limits anterior gliding of the atlas during flexion. It usually fails suddenly when either rapid or slow-loading forces are applied. After the transverse ligament ruptures the support provided by the remaining C0-C2 ligaments, which stretch with relative ease, is inadequate to prevent significant displacement of C1 and C2.

It had been long suggested that the axial rotation of the atlantoaxial joint is limited to one side by tightening of the alar ligament of the other side. But Crisco et al⁵ and Panjabi et al¹ found a single alar ligament transection resulted in significant increase in axial rotations to both sides. This indicates that the two alar ligaments work in unison. The functional loss of alar ligaments implies a potential for rotatory instability of the C1-C2 joint, consequently, keeping spinal cord and vertebral artery from rotatory injury.

The C1-C2 joint structure has considerable stable function and a potential for rotatory instability. Axial rotation (one side ) at the C1-C2 joint was 47.93 degrees with the neutral zone constituting 75.99% of this motion. Congenital anomalies (atlanto-occipital fusion, dentis deformity, congenital loss of transverse ligament of atlas) result in displacement of C1 and C2 besides trauma, infection (pharyngealthroatitis, rheumatism).⁶ Zhou et al⁷ reported that there were 19 cases of C0-C1 fusion and 13 cases of C2-C3 fusion in 23 cases of congenital atlanto-axial dislocation. It implies that the atlanto-axial joint endures more stress due to the C0-C1 joint and the C2-C3 joint motion losing. Thereby, the tight of transverse ligament and alar ligament is increased, as a result, instability of the C1-C2 joint either subluxation or dislocaton occurs.

We observed that the main motions of flexion and extension at the C1-C2 joint companied with axial rotation. The ranges of the main motions were close to the coupled motion. This feature was confirmed by Dvorak⁸ reporting functional x-rays examination of the normal cervical spine too. Meanwhile, we also found coupled rotation with the lateral bending, and the ranges of the coupled rotation were about three times of the main motion. Worne regarded the C1-C2 joint as ball and socket joint, which may explain why coupled rotation is accompanied by the lateral bending. Therefore, when the C1-C2 joint was fused for the treatment of atlantoaxial anterior dislocation, not only anterior gliding of the atlas but also rotation instability of the C1-C2 joint should be avoided in this way, bony fusion can be gained between bone blocks and C1-C2.

REFERENCES
[1] Panjabi MM, Dvorak J, Crisco JJ, et al. Effects of alar ligament transection on upper cervical spine rotation. J Orthop Res 1991; 9∶584.
[2] Panjabi MM, Dvorak J, Duranceau J, et al. Three-dimensional movements of the upper cervical spine. Spine 1988; 13∶726.
[3] Oxland TR, Lin RM, Panjabi MM. Three-dimensional mechanical properties of the thoracolumbar junction. J Orthop Res 1992; 10∶573.
[4] Dickman CA, Sonntag VKH, Browner CM, et al. Transverse atlantal ligament imaging. J Neurosurg 1991; 75∶221.
[5] Crisco JJ, Panjabi MM, Dvorak J. A model of the alar ligaments of the upper cervical spine in axial rotation. J Biomech 1991; 24∶607.
[6] General Hospital of CPLA: Practical Neurosurgery. 10th ed. Beijing: CPLA Soldier, 1978∶ 736.
[7] Zhou DB, Duan GS, Zhang J, et al. Diagnosis and treatment of congenital atlantoaxial dislocation. Chin J Surg 1991; 29∶734.
[8] Dvorak J, Panjabi MM, Novotny JE, et al. In vivo flexion/extension of the normal cervical spine. J Orthop Res 1991; 9∶828. (LU Yi-sheng(卢一生), JIA Lian-shun(贾连顺),DING Zhu-quan(丁祖泉), DAI Li-yang(戴力扬))