Poster Presentation 50th International Society for the Study of the Lumbar Spine Annual Meeting 2024

INTRODUCTION 

Freehand pedicle screw insertion is heavily reliant on surgical expertise in conjunction with sense perception. Auditory feedback, in the form of distinct vibration patterns created during pilot hole drilling, is used to deduce whether the drill bit has perforated cortical walls [1]. Tactile feedback during screw insertion, which is the feeling of torque in the hands, is used to determine if screws have breached cortical bone [2]. The danger of breaching is iatrogenic injury and an increased risk of screw loosening and non-union. A large proportion of spinal surgeons do not perform preoperative bone mineral density scans [3]. Therefore, in clinical settings, surgeons rely on auditory and tactile perceptions to identify breach, to judge the quality of bone under fixation and extent of screw purchase. This study investigated breach diagnosis using acoustic emission (AE) and torsional force (TF) profiling. Furthermore, it determined how breaches influence fixation stability, defined using pullout factors of maximum force (F_max), stiffness (S), energy (E) and displacement (d).

METHODS

In vitro experimental testing was conducted on 58 ovine lumbar vertebrae (L1 – L6), randomly subdivided into three trajectory groups: Normal Insertion (n=26), Lateral Breach (n=16) and Medial Breach (n=16). Pilot holes were drilled at 1250 rpm into bones utilising an orthopedic drill and 2.5 mm drill bit. AE was captured at 48 kHz by a sound sensor module and maximum AE (AE_max) extracted. Self-tapping pedicle screws (4.5 x 32 mm), possessing singular threading, were inserted with a custom rig at 6.0 rpm following predrilled trajectories, whilst TF was sampled at 20 Hz. Maximum torsional force (TF_max) was extracted. A universal electromechanical test machine destructively pulled screws out of vertebrae. Screws were evulsed at 5 mm/min, concurrently force and displacement were recorded at 50 Hz. Thereafter, pullout factors were calculated. Data were collated and stratified by trajectory group. All factors were subjected to separate, one-way ANOVA analyses, followed by Tukey-Kramer or Dunn-Bonferroni post-hoc tests, to investigate differences between breached and non-breached trajectories.

RESULTS

The overall effect of trajectory was significant for TF_max (p = 0.02), however not for AE_max (p = 0.56). Trajectory was significant for all pullout test factors (p < 0.05). Post-hoc tests indicated significant decreases in all pullout factors for medial breach relative to normal insertion, including inferior pullout force, energy absorbed, structural stiffness and displacement range (Figure 1). Similarly, compared to normal insertion, lateral breach resulted in significantly less energy absorption before failure. Medial breach produced significantly lower pullout force and energy absorption than lateral breach.

DISCUSSION

Means of TF_max differed by trajectory and AE_max did not, however individual AE and TF curves revealed valuable information about the tool path through bone. Changes in AE and TF responses occurred during breach; consequently, these variables could be used intraoperatively to rapidly detect breach. Yet, they remain largely unexplored and are not currently measured in theatre. Using these parameters for continuous monitoring of cortical walls may reduce the number of dangerous breaches, which were herein demonstrated to significantly decrease bone-screw construct mechanical integrity.