Part-1: Comparison with stand-alone implants and conventional fixation method

Recently, surgical trends have shifted using minimally invasive techniques through alternative approaches and reduced loads of fixation devices to mitigate the issues of concern.

Examples of such techniques are stand-alone cages coupled with self-stabilizing mechanisms, such as the Latero plate, SynFix screw, and Stabilis cylinder, which eliminate the need of posterior fixation.

5.1.1. ROM

In this study, the results of ROM control showed that the A+P model had the best overall performance and the Stabilis model had the weakest ROM control. Both Latero and SynFix models were shown to have <(-50%) of ROM control in all motions except for that of the Latero model, which demonstrated a week control of right lateral bending (-37%). The asymmetrical design of the Latero cage and its instrumentation in the left side of interspace may explain the

67

different left and right bending behaviors. In practice, this discrepancy might be compensated for by adding bone grafts or elongating the cage to cover the right side of the interspace. Various biomechanical studies have shown that stand-alone cages lack control of extension and axial rotation but not of flexion and lateral bending [81]. The Latero model (-92%) had comparable control in extension as the A+P model (-94%) and was significantly superior to the SynFix model (-51%). This could be explained by the force vector of extension which was directly perpendicular to the Latero plate. This also accounts for the excellent control of extension and axial rotation by the Latero plate in comparison to the other counterparts (Figure 4. 1).

Although the rigidity of the screw-bone interface may be enhanced by injecting cement, there is the possibility of cement-related hazards. In Stabilis, the bony endplates might be compromised by threading in the cylindrical cage, and a higher incidence of cage subsidence than in other cages has been reported [72]. At the adjacent levels, their segmental ROMs were indeed changed when compared with the intact state. However, the changes were small among these four models, even in the stiffest A+P model which is known to have a higher incidence of adjacent segment disease at long-term follow up. This finite element study cannot reflect the consequences of clinical cyclic loading. It can only reveal the ROM changes of adjacent levels.

The varying design concepts of the integrated parts in the self-stabilizing cages not only affect their ROM control, but also influence the stress distribution on the device and the adjacent tissues. In this study, the stress distribution on the Latero plate was much lower and more evenly distributed than that of the SynFix screws (Figure 4. 5). The SynFix screws sustained 3.75 to 9.57 times greater stresses than the Latero plate, with the stress concentrated at the cage-screw junctions. For the Latero and SynFix models, the difference in modes and amounts of distributed stress might be explained by the different configurations of the stabilizing components: sizes, contact areas, and the distance to the center of rotation. In cyclic loading, the Latero plate might be less likely to have fatigue failure than the SynFix screws. The cylindrical cage of the Stabilis model sustained the lowest stress as a result of its relatively

68

suboptimal ROM control.

5.1.2. von-Mises stress at the vertebra

Stress distribution on the vertebra-cage interface represents vertical compression force exerted on the bony endplate. The Latero model exerted the highest stress on the L4 endplate compared to the other models (Figure 5. 1). This might be due to the stress not being fully shielded by the intervertebral plate and being redistributed by the plate and transmitted onto the L4 endplate. Comparatively, the stress was shielded by the screws of the SynFix and A+P models and thus less stress was distributed on the endplate. In the Stabilis model, because of its relatively poor ROM control, the distributed stress on the endplate was less than in the other three models. There were two biomechanical implications in the Latero model, which had higher vertical load on the L4 endplate than the other counterparts. The first implication indicates that higher vertebral stress may increase the incidence of cage subsidence particularly in suboptimal bone density. On the other hand, unshielded vertical load may be beneficial for graft consolidation according to Wolff's law.

69

Figure 5. 1: Distribution of endplate stress on the upper surface of the L4 vertebra for all models.

(A) Flexion. (B) Extension. (C) Left lateral bending. (D) Left axial rotation.

5.1.3. Facet contact force

After interbody fusion, abnormally high transmission of loads to the facet joints may ultimately result in arthritic changes. At the surgical level, when comparing the three stand-alone cage models, the Latero model was shown to have the lowest values of facet contact force in bilateral rotation and was near absent in extension (Figure 4. 4). The facet joints at the surgical level seemed to be relatively well protected in the Latero model in comparison to the SynFix and Stabilis models. At the adjacent levels, there were only few differences of facet contact force among all four instrumented models. This may indicates that the reasons for the degenerative facet joints at the adjacent levels might not be attributed to the facet contact force

70

but other clinical factors.

5.1.4. Stress distribution at the annulus

Distribution of annulus stress can provide the clinical implication that higher stress may result in annulus disruption and disc herniation. At the surgical level for the four instrumented models, the distribution of annulus stress was well correlated inversely to their ROM control (Figure 4. 3). At right lateral bending, the highest annulus stress was found in the Latero model, where it was concentrated at the right annulus. This corresponded to the relatively inferior control of the Latero due to its asymmetrical design. The posterior and left sides of annulus in the Stabilis model sustained the highest stress at extension and lateral bending, which manifested in its inferior control of those moments.

5.1.5. von-Mises stress at the cages

For the stresses of integrated parts, the locking screws of the SynFix model experienced extreme values under extension and lateral bending, and the stress concentrated at the connection area between the cage frame and the screws. In contrast, the thread of Stabilis had relatively lower stress under extension and lateral bending. In Latero and SynFix models, the maximum stresses of the implants occurred on the side plate and locking screws, respectively.

However, the stress on the locking screws is larger than that on the side plate. The stress on the locking screws is even 9.57 times of the stress on the side plate under extension. So, the fracture may occurred in the position of the locking screws. The fatigue tests with detailed boundary conditions can be simulated in the future.