By: 24 April 2020
Investigation of screw fixation in cervical fusion using artificial bone modelling

The Orthopaedic Innovation Centre in Canada performs biomechanical testing to determine optimal lateral mass screw fixation method in cervical spinal fusion

Cervical spinal fusion is a surgical procedure that joins two or more adjacent vertebrae in the neck to treat severe pain, disorder, trauma injury, and other spinal illnesses. A posterior fusion of the subaxial cervical vertebrae is routinely performed with screws implanted into the lateral mass regions of these vertebrae and joined with spine rods. The success of interbody fusion is greatly dependent on the level of screw fixation. Screw fixation can be affected intraoperatively by screw design, path of trajectory, purchase depth, and bone quality. However, achieving optimal screw fixation in cervical spine presents several challenges due to the small size of the lateral mass and its anatomic proximity to the vertebral artery and nerve roots. The importance of screw fixation is critical, as screw-bone failure may result in catastrophic neurologic injury and major spinal revision surgery.

Lateral mass screw fixation has been investigated by the biomedical engineers at the Orthopaedic Innovation Centre (Winnipeg, Canada) and the orthopaedic surgeons and residents from the Department of Surgery at the Health Sciences Centre and University of Manitoba (Winnipeg, Canada). They have combined orthopaedic practice with biomechanical testing to determine optimal screw fixation in subaxial cervical spinal fusion. Screw design, trajectory angle, and depth of fixation were investigated.

Research testing involved the use of artificial bone modelling (figure 1) to simulate clinical conditions (figure 2). Lateral mass screws were tested in bi-cortical cancellous filled phantom bone blocks with material characteristics similar to cervical spine. Screw fixation was determined through mechanical testing (figure 3a) by measurement of maximum axial load (N) to pull-out failure, defined as screw pull-out from bone or bone fracture (figure 3b).

Figure 1: Mechanical axial pull-out test set-up to measure maximum load to screw pull-out failure.

Figure 2. Clinical representation of lateral mass screw trajectory as demonstrated in figure 1.

Figure 3a. Mechanical axial pull-out test set-up to measure maximum load to screw pull-out failure.

Figure 3b. Maximum load-to-failure of lateral mass screw, resulting in bone block fracture.

Cortical and cancellous lateral mass screws were tested for fixation in Roy-Camille and Magerl trajectories for three purchase depths. Cortical screws are designed to have small thread pitch, while cancellous screws have larger thread pitch. Pilot holes were drilled into the bone blocks at either a Roy-Camille trajectory described as a lateral angulation of 10° from the sagittal plane and a sagittal inclination of 90° to the lateral mass surface, or a Magerl trajectory described as a lateral angulation of 25° from the sagittal plane and a sagittal inclination parallel to the adjacent facet joints (45° to the lateral mass surface) (Joaquim et al., Global Spine J 2018). Screws were seated into the blocks using specialised orthopaedic instrumentation to either bi-cortical purchase, uni-cortical purchase, or bi-cortical then backed out to a uni-cortical purchase (figure 4).

Figure 4. Test preparation showing lateral mass screw insertion into bi-cortical artificial bone blocks.

Six replicates from each group were tested and statistical comparisons (p<0.05) were made between groups. Screw length (18mm) and diameter (3.5mm) were controlled. However, thread counts differed between screw designs. Therefore, it was hypothesised that fixation between screw designs would differ. Similarly, trajectory angles were expected to yield different pull-out loads. Furthermore, fixation of screws inserted to a bi-cortical purchase was predicted to be greater than a uni-cortical purchase, but the interest was by how much?

Results indicated a variation in fixation between screw types that was dependent on trajectory and purchase depth. Screws implanted through a close-to-perpendicular trajectory (Roy-Camille) indicated significantly greater fixation for the cortical screw than the cancellous screw, for all purchase depths. This result may be attributed to a smaller thread pitch on the cortical design (15 threads over 18mm) than the cancellous design (9 threads over 18mm), therefore providing a greater surface area for fixation within bone.

On the other hand, when tested at a higher angulation and inclination (Magerl), the cancellous and cortical screws performed with similar fixation in bi-cortical and uni-cortical purchase. However, significantly better fixation was seen in the cancellous screws after backing out from bi-cortical to uni-cortical purchase.

Screw trajectory revealed a large influence on fixation. The Roy-Camille trajectory demonstrated significantly stronger fixation than the Magerl trajectory for both screw types and purchase depths. In fact, the Roy-Camille technique showed a 45-65 per cent increase in cancellous screw pull-out strength and an 80-135 per cent increase in cortical screw pull-out strength from those with a Magerl trajectory.

This large difference may be explained by the level of bending created from the axial pull-out load direction and screw inclination angle. The findings support that lateral mass screws implanted through a higher inclination angle (Magerl trajectory) cause bone fracture at lower loads than when implanted through an angle closer to the pull-out direction (Roy-Camille trajectory).

A bi-cortical purchase gave significantly greater fixation than a uni-cortical purchase. Specifically, a bi-cortical purchase increased load-to-failure by approximately 15-30 per cent from a uni-cortical purchase for both screw types and trajectories. Ultimately, deeper screw penetration increases the screw-to-bone contact surface area, corresponding to improved fixation.

It was anticipated that backing out screws from bi-cortical to uni-cortical would lessen screw fixation from if they were seated to a uni-cortical depth initially. While this effect showed a reduction in fixation, it was only significant for cancellous screws implanted through a Roy-Camille trajectory (10 per cent reduction), and for cortical screws implanted through a Magerl trajectory (25 per cent reduction).

Overall, the greatest lateral mass screw fixation was found with a cortical screw design through a Roy-Camille trajectory at a bi-cortical purchase (mean = 825N). On the other hand, the weakest fixation was also seen with a cortical screw but implanted along the Magerl path and backed out from a bi-cortical to a uni-cortical purchase (mean = 285N).

In summary, lateral mass screw fixation investigated through artificial bone modelling has demonstrated the importance of screw design, screw trajectory, and purchase depth for subaxial cervical spinal fusion. Cortical screws were shown to provide better fixation than cancellous screws as a result of tighter thread pitch. Screws were able to withstand higher pull-out loads when implanted closer to perpendicular to the lateral mass surface. Lastly, deeper screw implantation improved fixation, despite the trajectory or screw design.

While screw fixation has been measured through mechanical testing, there were some limitations. Fixation was calculated as the maximum pull-out load directed perpendicular to the lateral mass surface, demonstrated by the test blocks. Meanwhile, fusion hardware is subject to loads from all directions clinically. Therefore, screws may become loose as a result and fixation may vary. Additionally, osteoporotic bone may further decrease implant fixation.

Although the effect of cyclic loosening on screw fixation was not explored, intraoperative subaxial cervical spine screw fixation of various techniques has been determined. Furthermore, the significance of screw type, trajectory, and purchase depth in posterior cervical fusion has been outlined to reduce the likelihood of bone fracture and catastrophic failure.

The Orthopaedic Innovation Centre continues to combine biomechanical testing with orthopaedic practice to improve the healthcare community and product development. All collaborators on this topic would like to acknowledge the Alexander Gibson Fund at the University of Manitoba for funding this research. The researchers and engineers have no relevant industry disclosures.



Joaquim, AF., Mudo, ML., Tan, LA., Riew, KD. Posterior subaxial cervical spine screw fixation: a review of techniques. Global Spine J, 2018. Vol (8):7 p:751-760.



Sara Parasin is a biomedical engineer at the Orthopaedic Innovation Centre (Winnipeg, Manitoba) who dedicates her time to orthopaedic research studies, engineering testing, and explant retrieval analysis. Sara leads the biomechanical testing and clinical research departments related to the spine.