(B-45) Numerical Simulation of Respiratory Airflow in Central Airways with Elastic Boundary Conditions

Introduction:: Advancements in computational processing and methodologies have allowed numerical simulation to become a powerful tool and an effective approach for the prediction of respiratory and cardiovascular phenomena for bioengineering and clinical applications. Due to the complexity of biological systems, it is challenging for computational models to maintain physiologic fidelity and yield realistic predictions over a wide range of normal and pathologic conditions. In this study, we simulated airflow within human central lung airways, from the trachea to the fourth generation, with elastic boundary conditions for normal spontaneous breathing, conventional mechanical ventilation (CMV), and multi-frequency ventilation (MFV). By using forced gas flows at the airway opening of a computational lung model, dynamic pressures at various airway segments were predicted based on the interactions of internal flow with downstream elastances, and peripheral airway resistances.

Materials and Methods:: A three-dimensional (3D) airway tree was obtained from the segmentation of a thoracic image of a healthy adult subject. It was then truncated from the proximal third of the trachea to half of the 4^th generation airway segments. Due to asymmetric branching, the resulting airway tree consisted of fourteen peripheral outlets. Sinusoidal, CMV, and MFV airflow waveforms were imposed as boundary conditions at the airway opening. The sinusoidal waveform was used to simulate normal spontaneous breathing with a frequency of 0.333 Hz and tidal volume (V_T) of 500 mL. The CMV waveform was directly sampled from an adult patient under general anesthesia and was adjusted to zero-mean flow by offsetting the positive and negative flow volumes. The MFV waveform was simulated by applying a small sinusoidal oscillation at 5 Hz and 100 mL V_T to a fundamental sinusoidal waveform of 0.4 Hz and 500 mL V_T. Dynamic pressure variations at each of the fourteen outlets of the 4^th generation airway segments were induced by the effects of downstream parenchymal tissue elastance, airway resistance, and gas inertia, and were applied as boundary conditions. All the values of downstream airway resistance and air inertance were internally validated by inverse modeling regression [1]. The airway walls were assumed to be rigid, with a no-slip boundary condition. Turbulent kinetic energy and dissipation of the internal airflow was solved assuming a k-ω SST turbulence model. The 3D Navier-Stokes equations were solved using a Pressure-Implicit with Splitting of Operators (PISO) algorithm within the airway tree.


Results, Conclusions, and Discussions:: The pressure variations at the 14 outlets showed similarity to each other in terms of magnitude and phase, regardless of airway opening waveform. During normal spontaneous breathing, elastic pressure dominates the outlet pressures at slightly over 4 cmH₂O. The CMV waveform yielded outlet pressures of approximately 5 cmH₂O, which were also dominated by elastic pressure. For MFV, peak outlet pressures were approximately 7 cmH₂O, which were comprised predominately of elastic and resistive components. Moreover, 90-degree phase differences between these pressure components were also observed. For all three waveforms, the distributions of wall shear stress were dramatically different during inspiration and expiration. Regions of airway bifurcation experienced the highest wall shear stresses during inspiration and the lowest shear stresses during expiration. MFV induced lower peak-to-peak shear stress on the sidewalls compared to spontaneous breathing or CMV, suggesting its potential to reduce airway epithelial injury.
In this study, we simulated flow phenomena within human central airways during spontaneous breathing, CMV, and MFV. Our simulations offer an effective way to reveal flow, pressure, and shear stress phenomena for specific airway regions and across various ventilation modalities. These results highlight the importance of applying realistic boundary conditions in such numerical airflow simulations.


Acknowledgements (Optional): : Supported in part by the Office of the Assistant Secretary of Defense for Health Affairs, through the Peer Reviewed Medical Research Program under Award Number W81XWH-21-1-0507 and W911NF-23-1-0004. Opinions, interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the Department of Defense. The authors appreciate Dr. Adam Sonnenberg from UIUC Carle Illinois College of Medicine for providing the 3D airway cast.


References (Optional): : [1]Kaczka DW, Barnas GM, Suki B, Lutchen KR. Assessment of time-domain analyses for estimation of low-frequency respiratory mechanical properties and impedance spectra. Ann Biomed Eng 1995; 23:135-151.