Acoustic emission-based diagnostic tool for temporomandibular joint implant fixation screw loosening

Alloplastic temporomandibular joint replacement (TMJR) implants have become an acceptable clinical solution for end-stage temporomandibular joint (TMJ) pathology. However, early screw fixation loosening can be a concern. Acoustic emission (AE) has been reported to successfully detect the early deformation of hip implants. We hypothesize that AE can also be utilized to predict early TMJR implant failure due to fixation screw loosening.

The experiments were conducted on a custom-built TMJ simulator equipped with AE, bite force and displacement sensors, along with a patient-fitted Stryker/TMJ Concepts (Ventura, CA) implant containing a mandibular ramus component with eight fixation screw sites. Four screw loosening sites (S1, S2, S7, and S8) were selected for the study based on prior studies⁴, while the rest of the fixation screws remained tight. Four screw fixation study groups were established to simulate different TMJ implant conditions (e.g., fully fixed, partially fixed, and failed), (i) All fixed screws as the control group, (ii) S1 and S2 loosened by 360°, (iii) S7 and S8 loosened by 360°, and (iv) all four screws loosened. The TMJ simulator moved at 1 Hz with a 10N bite force throughout the experiment.

This paper reports the results of the three phases of this study. In Phase 1, the pilot study found a high correlation between the increase in AE energy and the amount of loosened fixation screws. In Phase 2, the simulator applied a masticatory force of 10N to the TMJR components implanted on the model. The displacement was approximately 22 mm ± 2 %. The friction coefficient was calculated to be approximately 0.2 ± 0.05 based on the friction force for all groups⁴. AE signals were collected for hit-driven data and amplitude for all groups, where Group (ii) exhibited higher peaks than (i), Group (iii) showed more peaks than (i) but lower than Group (ii), and Group (iv) demonstrated the hits similar to Group (ii). In Phase 1, a direct correlation has been made with the AE data and loosened and tightened screws. In Phase 2, the sensors were used in the mandible of the skull to detect the AE stress waves produced by the tightened and loose screws. The data was recorded, but it was not able to distinguish between the loosened and fixed screws. In Phase 3, the sensors were placed directly on the screws of the TMJ implant when they were tightened or loosened. During this phase, distinct differences have been observed between the loosened and tightened conditions. Therefore, this demonstrated that the tightening activity of the other screws prevents us from seeing the AE energy for the fixed screws when we look at the system as a whole.

The findings show that the AE sensors could detect the stress wave signals when the AE sensors were placed directly on the implant. The differences in signals are likely due to the mechanical and tribological interactions at the screw-bone interface. The custom-built TMJ simulator and the results obtained thus far might lead to further studies of fixation screw loosening in TMJ implants.