Fluid-Structure Interaction Analysis in Human Upper Airways to Understand Sleep Apnea.

Authors: Mylavarapu, G., Mihaescu, M, Gutmark, E., Murugappan, S., Kalra, M.
Document Type: Article
Pubstate: Published
Journal: 48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition.
Volume: AIAA 2010-1264   10 pages
Year: 2010


Sleep apnea is characterized by partial or complete obstruction of upper airway during sleep. Existing clinical therapies are not fool-proof. Understanding human upper airway mechanisms with flow and structure dynamics is hence, essential for designing better clinical therapies. In this study, two dimensional Fluid-Structure Interaction (FSI) simulations were carried on human upper airway models using ADINA 8.4, a finite element code. Baseline model (B) is reconstructed from a mid-sagittal Magnetic Resonance Image (MRI) of an adolescent. In-house developed MATLAB code is used to de-feature various structures on the MRI scan. Vertices information from MATLAB code is then imported into Gambit 2.2, where edges are reconstructed and exported in IGES format to ADINA. Fluid and Solid surfaces are reconstructed and discretized in ADINA. Flow was assumed laminar and incompressible. Plain Strain hypothesis and small strains were assumed with solid domain. Appropriate boundary conditions with wall and fluid structure interfaces were defined in both fluid and solid domains. Pressure drop across airway is varied incrementally with a user defined time function. Airway wall displacement and flow features are obtained. The displacement of tip of soft palate and critical closing pressures required for partial or complete closure of upper airways are computed and compared across different cases. For comparative studies between a normal, narrower and corrected airway models (A, B, C), Model B is reconstructed from Model A by moving its posterior airway wall in anterior direction and Model C is reconstructed from Model B by excising the length of soft palate by 30%. It was observed that airway is more susceptible to collapse during expiration phase than inspiration. Increasing the stiffness of soft palate as in a palatal implant clinical therapy showed significant improvement in airway potency as demonstrated in Models A and B. Model B demonstrates a case of complete obstruction for significantly lower closing pressures when compared to Model A. Model A demonstrated hypo-apnea or partial obstruction of airway. Model C with excised soft palate is less susceptible to airway collapse when compared to Model B. This study is step forward to our previous studies on upper airways where airway walls were assumed static.