An ovine knee simulator: description and proof of concept.

ABSTRACT

Aims: The ovine stifle is an established model for evaluation of knee treatments, such as meniscus replacement. This study introduces a novel ovine gait simulator for pre-testing of surgical treatments prior to in vivo animal trials. Furthermore, we describe a pilot study that assessed gait kinematics and contact pressures of native ovine stifle joints and those implanted with a novel fiber-matrix reinforced polyvinyl alcohol-polyethylene glycol (PVA-PEG) hydrogel meniscus to illustrate the efficacy of the simulator. Methods: The gait simulator controlled femoral flexion-extension and applied a 980N axial contact force to the distal tibia, whose movement was guided by the natural ligaments. Five right ovine stifle joints were implanted with a PVA-PEG total medial meniscus replacement, fixed to the tibia via transosseous tunnels and interference screws. Six intact and five implanted right ovine stifle joints were tested for 500 k gait cycles at 1.55 Hz. Implanted stifle joint contact pressures and kinematics in the simulator were compared to the intact group. Contact pressures were measured at 55° flexion using pressure sensitive film inserted sub-meniscally. 3D kinematics were measured optically across two 30-s captures. Results: Peak contact pressures in intact stifles were 3.6 ± 1.0 MPa and 6.0 ± 2.1 MPa in the medial and lateral condyles (p < 0.05) and did not differ significantly from previous studies (p > 0.4). Medial peak implanted pressures were 4.3 ± 2.2 MPa (p > 0.4 versus intact), while lateral peak pressures (9.4 ± 0.8 MPa) were raised post medial compartment implantation (p < 0.01). The range of motion for intact joints was flexion/extension 37° ± 1°, varus/valgus 1° ± 1°, external/internal rotation 5° ± 3°, lateral/medial translation 2 ± 1 mm, anterior/posterior translation 3 ± 1 mm and distraction/compression 1 ± 1 mm. Ovine joint kinematics in the simulator did not differ significantly from published in vivo data for the intact group, and the intact and implanted groups were comparable (p > 0.01), except for in distraction-compression (p < 0.01). Conclusion: These findings show correspondence of the ovine simulator kinematics with in vivo gait parameters. The efficacy of the simulator to evaluate novel treatments was demonstrated by implanting a PVA-PEG hydrogel medial meniscal replacement, which restored the medial peak contact pressures but not lateral. This novel simulator may enable future work on the development of surgical procedures, derisking subsequent work in live animals.