Development and validation of a computational finite element model of the rabbit upper airway : simulations of mandibular advancement and tracheal displacement

The mechanisms leading to upper airway (UA) collapse during sleep are complex and poorly understood. We have previously developed an anaesthetized rabbit model for studying UA physiology. Based on this body of physiological data, we aimed to develop and validate a two-dimensional (2D) computational finite element (FEM) of the passive rabbit UA and peripharyngeal tissues.Model geometry was reconstructed from a mid-sagittal CT image of a representative NZ White rabbit, which included major soft (tongue, soft palate, constrictor muscles), cartilaginous (epiglottis, thyroid cartilage) and bony pharyngeal tissues (mandible, hard palate, hyoid bone). Other UA muscles were modeled as linear elastic connections. Initial boundary and contact definitions were defined from anatomy and material properties derived from the literature. Model parameters were optimized to physiological data sets associated with mandibular advancement (MA) and caudal tracheal displacement (TD), including hyoid displacement, which featured with both applied loads. The model was then validated against independent data sets involving combined MA and TD. Model outputs included UA lumen geometry, peripharyngeal tissue displacement, stress and strain distributions.Simulated MA and TD resulted in UA enlargement and non-uniform increases in tissue displacement, stress and strain. Model predictions closely agreed with experimental data for individually applied MA, TD and their combination.We have developed and validated an FEM of the rabbit UA which predicts UA geometry and peripharyngeal tissue mechanical changes associated with interventions known to improve UA patency. The model has the potential to advance our understanding of UA physiology and peripharyngeal tissue mechanics.15 page(s