Flow in a terminal alveolar sac model with expanding walls using computational fluid dynamics

Abstract

Particles depositing on alveolar walls is of concern due to their potential to migrate through the blood-gas barrier. Whole-lung dosimetry models do not account for the flow field inside the alveoli and therefore may not accurately describe alveolar deposition. Studies that quantify the flow patterns in realistic geometries are limited and results inconsistent. This study aims to better understand the fluid characteristics in the terminal air sacs; specifically, alveolar mouth to depth flow rate ratio, penetration depth of residual air, and diffusive versus convective particle motion. A terminating alveolar sac with expanding alveolar walls was constructed using 13 truncated sphere-shaped alveoli, with dimensions consistent with published morphometry data. The flow field was governed by a measured in vivo breathing curve for normal volumes over periods of 2 and 4 seconds, analyzed numerically and compared to previous literature. Recirculation was not present, consistent with prior studies. Flow rate ratios (0.18–0.36) were within the range (0.057–1) previously reported. Penetration depths were less than 33% into the air sac during inhalation, decreasing in length for air inside the sac to zero near the wall. Péclet numbers indicated diffusion dominated flow for all submicron-sized particles. However, convection was significant at the duct entrance for particles >0.5 micron and inside the sac for particles >1 micron. Wall motion induced convection may not always be negligible, and if neglected could affect the accuracy of deposition predictions for certain particle sizes and flow conditions.

Keywords::

• Lung
• Alveolar Sac
• CFD
• Expanding
• Pulmonary
• Convective Flow
• Peclet

Acknowledgments

The authors would like to thank Dr. Kambiz Nazridoust (Division of Computational Biology, Hamner Institutes for Health Sciences) for contribution in developing the computer code used for simulating breathing and Ryan Lewis (Application Consultant Specialist, Research Computing, RIT) and Paul Mezzanini (Senior Systems Administrator, Research Computing, RIT) for providing support to their various computational resources.

Declaration of interest

This work was completed with the support of the American Cancer Society (RSG-05-021-01-CNE).