Abstract Rationale Mandibular Advancement Devices (MADs) are primary interventions for Obstructive Sleep Apnea (OSAHS), yet the clinical efficacy is frequently undermined by suboptimal advancement titration. The fundamental biomechanical relationship between mandibular protrusion and upper airway deformation remains uncertain. This study aimed to define this relationship and identify an optimal therapeutic threshold using a high-fidelity multiphysics modeling approach. Methods An integrated patient-specific finite element (FE) models, derived from Cone-Beam Computed Tomography (CBCT), with measured mandibular kinematic trajectories is modeled. Fluid-Solid Coupling (FSI) analysis was employed to resolve airflow dynamics and airway wall stress at varying protrusion levels. This in silico framework was rigorously validated against a biomimetic 3D-printed pharyngeal phantom, fabricated using ex-vivo tissue mechanical properties. Neuromuscular constraints were concurrently assessed via surface electromyography (EMG). Results Results revealed that advancement non-linearly increases airway volume, which plateaus beyond 75% of maximum protrusion, mitigating pathological turbulence in the retro-palatal and retro-glossal regions. FSI analysis identified the retro-epiglottic region as the principal site of airflow obstruction. Critically, we identified an optimal biomechanical window—at 60-80% of maximum protrusion—where airway compliance is maximized and airflow resistance is minimized. However, EMG data revealed a critical neuromuscular constraint: protrusion approaching the maximum limit induces significant masseter and mylohyoid activity, indicating a high risk of muscle fatigue. Conclusion The optimal therapeutic window for MADs is a trade-off between maximizing airway patency and minimizing neuromuscular strain. Findings establish that clinical protocols must avoid maximum protrusion to prevent muscle fatigue, a primary driver of treatment non-compliance. We leveraged these data-driven parameters to design and successfully validate a novel adjustable MAD, demonstrating the feasibility of this mechanistic approach for personalized treatment. This abstract is funded by: None
Qu et al. (Fri,) studied this question.