Superior biomechanical stability of pedicle screws compared to lateral mass screws: recommendations for bicortical positioning and enhancing bone contact in geriatric C1 vertebrae

Abstract

Background: In atlantoaxial instabilities, posterior C1/C2 fusion using lateral mass screws (LMS) or pedicle screws (PS) in a mono- or bicortical position in the atlas is a typical treatment. The bone microstructure and positioning of the screw trajectories appear to be of significant relevance for stability.

Purpose: The aim of this study was a comparative analysis of the mechanical durability of screw fixation concerning microstructural characteristics of the trajectories of LMS and PS in mono- and bicortical position.

Methods: Human C1 from geriatric body donors (n = 28; 50% female, age 80.8 ± 13.9 years) were collected and characterized based on their bone microstructure. Additionally, the mechanical stability of LMS and PS fixation in mono- and bicortical positioning was tested by mechanical loading. High-resolution quantitative computed tomography was used to analyze the bone microstructure of cylinders corresponding to the trajectories of PS and LMS in mono- and bicortical locations in each C1. After instrumentation with both screw types and types of fixation, the mechanical stability was tested by increased cyclic loading in cranio-caudal direction.

Results: Trajectories of PS presented with more bone volume and a higher contact length to cortical bone. Simultaneously, a higher number of cycles and a higher maximum force was needed to loosen PS compared to LMS, while the loose by torque at the experiment end was still greater in PS. Differences between mono- and bicortical positioning of PS and LMS have only been observed in the initial stiffness of screws. When comparing microstructural and mechanical properties, the cortical contact length and bone volume in screw trajectories were strongest associated with a high loose and cycle count.

Conclusions: This study suggests that mono- and bicortical positioning of PS is similarly efficient in creating a stable basis for screw fixation in the atlas. While PS are superior to LMS, the contact with cortical bone is of major relevance for a stable foundation.

Keywords: Bone microstructure; C1; Cervical spine surgery; Lateral mass screw; Mechanical testing; Pedicle screw.

Fig. 1

Representative scheme of an atlas and imaging of screw trajectories in mono- or bicortical screw fixation. (A) Three-dimensional renderings of an atlas with indicated cylindric trajectories of pedicle screws (PS) in red and lateral mass screws (LMS) in blue, as well as their entry points at the dorsal side of an atlas (right side). (B) Dorsal view of a representative atlas instrumented with a lateral mass screw on the left and a pedicle screw on the right. (C) Top view of an atlas obtained by digital contact radiography (DCR) illustrating monocortical (MC) and bicortical (BC) instrumentation.

Fig. 2

Scheme of the mechanical testing setup of LMS and PS fixation in atlases. Atlases were embedded in Technovit 3040 (yellow) and fixated by clamps attached to the testing machine’s table. The screw heads (LMS & PS) were stabilized with the corresponding metal rod template and grub screw. The metal pin was attached to a load cell by a clamping jaw and mechanically loaded by cyclic application of increasing force (F = -25 N ± 0.05 N/cycle). LMS: lateral mass screw, PS: pedicle screw, F₀: initial loading

Fig. 3

Microstructural characteristics of screw trajectories in the atlas. (A) A scheme illustrating the bone volumes measured by high-resolution quantitative computed tomography. In both the total (including the screw diameter and surrounding bone, and not the drill canal) and the surrounding bone volume, PS trajectories presented with a higher bone volume than LMS trajectories, while no difference between mono- and bicortical volumes were determined. (B) Representative lateral view and top-down view of a bicortical PS cylinder location to illustrate the cortical contact length passed (black). (C) PS presented with a longer cortical contact compared with LMS. Additionally, a trend towards a higher contact length in bicortical trajectories than in monocortical trajectories was apparent. (D) The total insertion length of the screws was higher in bicortical than in monocortical position, and in PS compared with LMS. (E) The estimated contact to the trabecular bone of PS was longer compared with LMS in both mono- and bicortical positions. (F) In PS a higher apparent volumetric bone mineral density (vBMD) than LMS was observed, while mono- and bicortical trajectories did not show differences

Fig. 4

Mechanical properties of PS and LMS fixation. (A) The displacement during testing is presented over cycles indicating a different behavior predominantly between PS (red) and LMS (blue), while differences between mono- (pointed) and bicortical (straight) fixation were less pronounced. The green line indicates the test endpoint at -1.3 mm. The initial displacement of PS was higher than LMS, while no significant difference between bicortical fixation to monocortical fixation was observed. (B) PS fixation presented with more cycles until the test end than LMS fixation. (C) Likewise, the maximum force of PS fixation was higher than LMS fixation. (D) After loading, PS presented with an elevated loose (torque) indicating a higher end stability than LMS. (E) Similar to the displacement, the stiffness primarily differed between PS and LMS fixation. (F) In the initial phase, PS fixation presented stiffer than LMS fixation, and the bicortical placement of PS also presented stiffer than monocortical placement. (G) At the endpoint, the stiffness of PS was higher than that of LMS

Fig. 5

Correlation analyses of microstructural parameters and mechanical properties. (A) Heatmap of correlation coefficients reflecting associations between the screw trajectory microstructure (i.e., total bone volume, volumetric bone mineral density, total insertion length, cortical contact length, estimated trabecular contact length) and demographics (i.e., age and BMI of the individuals) with experimentally determined mechanical properties of LMS and PS. The r coefficient as well as p-values are depicted. The color coding reflects the r coefficient. (B) The loose by torque of LMS and PS at the end of the experiments correlated strongest with the cortical contact length, while PS in mono- and bicortical fixation presented with a higher cortical contact length and loose values. (C) Similarly, a strong correlation of cycle count with the cortical contact length was indicated, while PS in mono- and bicortical fixation presented with a higher cortical contact length and cycle counts