Department of Biomedical Engineering, Technion - Israel Institute of Technology, Haifa 32000, Israel.
Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka 560-8531, Japan.
Eur J Pharm Sci. 2018 Feb 15;113:53-63. doi: 10.1016/j.ejps.2017.09.033. Epub 2017 Sep 24.
Computational Fluid Dynamics (CFD) have offered an attractive gateway to investigate in silico respiratory flows and aerosol transport in the depths of the lungs. Yet, not only do existing models lack sufficient anatomical realism in capturing the heterogeneity and morphometry of the acinar environment, numerical simulations have been widely restricted to domains capturing a mere few percent of a single acinus. Here, we present to the best of our knowledge the most detailed and comprehensive in silico simulations to date on the fate of aerosols in the acinar depths. Our heterogeneous acinar domains represent complete sub-acinar models (i.e. 1/8th of a full acinus) based on the recent algorithm of Koshiyama & Wada (2015), capturing statistics of human acinar morphometry (Ochs et al. 2004). Our simulations deliver high-resolution, 3D spatial-temporal data on aerosol transport and deposition, emphasizing how variances in acinar heterogeneity only play a minor role in determining general deposition outcomes. With such tools at hand, we revisit whole-lung deposition predictions (i.e. ICRP) based on past 1D lung models. While our findings under quiet breathing substantiate general deposition trends obtained with past predictions in the alveolar regions, we underscore how deposition fractions are anticipated to increase, in particular during deep inhalation. For such inhalation maneuver, our simulations support the notion of significantly augmented deposition for all aerosol sizes (0.005-5.0μm). Overall, our efforts not only help consolidate our mechanistic understanding of inhaled aerosol transport in the acinar depths but also continue to bridge the gap between "bottom-up" in silico models and regional deposition predictions from whole-lung models. Such quantifications provide what is deemed more accurate deposition predictions in morphometrically-faithful models and are particularly useful in assessing inhalation strategies for deep airway deposition (e.g. systemic delivery).
计算流体动力学(CFD)为研究肺部深处的呼吸流和气溶胶输运提供了一种有吸引力的途径。然而,现有的模型不仅在捕捉腺泡环境的异质性和形态方面缺乏足够的解剖学真实性,而且数值模拟也广泛局限于仅捕获单个腺泡的一小部分的域。在这里,我们根据 Koshiyama 和 Wada(2015 年)的最新算法,展示了迄今为止腺泡深处气溶胶命运的最详细和全面的计算模拟。我们的异质腺泡域代表了完整的亚腺泡模型(即整个腺泡的 1/8),捕获了人类腺泡形态的统计数据(Ochs 等人,2004 年)。我们的模拟提供了气溶胶输运和沉积的高分辨率、3D 时空数据,强调了腺泡异质性的变化如何仅在确定一般沉积结果方面发挥次要作用。有了这些工具,我们根据过去的 1D 肺模型重新审视了全肺沉积预测(即 ICRP)。虽然我们在安静呼吸下的发现证实了过去预测在肺泡区域获得的一般沉积趋势,但我们强调了沉积分数如何预计会增加,特别是在深吸气期间。对于这种吸气操作,我们的模拟支持所有气溶胶尺寸(0.005-5.0μm)的沉积显著增加的观点。总的来说,我们的努力不仅有助于巩固我们对腺泡深处吸入气溶胶输运的机制理解,而且继续弥合“自下而上”计算模型和全肺模型的区域沉积预测之间的差距。这种量化提供了在形态学上真实的模型中更准确的沉积预测,特别有助于评估深气道沉积的吸入策略(例如系统给药)。