A computational study of the respiratory airflow characteristics in normal and obstructed human airways

Obstructive lung diseases in the lower airways are a leading health concern worldwide. To improve our understanding of the pathophysiology of lower airways, we studied airflow characteristics in the lung between the 8th and the 14th generations using a three-dimensional computational fluid dynamics model, where we compared normal and obstructed airways for a range of breathing conditions. We employed a novel technique based on computing the Pearson׳s correlation coefficient to quantitatively characterize the differences in airflow patterns between the normal and obstructed airways. We found that the airflow patterns demonstrated clear differences between normal and diseased conditions for high expiratory flow rates (>2300 ml/s), but not for inspiratory flow rates. Moreover, airflow patterns subjected to filtering demonstrated higher sensitivity than airway resistance for differentiating normal and diseased conditions. Further, we showed that wall shear stresses were not only dependent on breathing rates, but also on the distribution of the obstructed sites in the lung: for the same degree of obstruction and breathing rate, we observed as much as two-fold differences in shear stresses. In contrast to previous studies that suggest increased wall shear stress due to obstructions as a possible damage mechanism for small airways, our model demonstrated that for flow rates corresponding to heavy activities, the wall shear stress in both normal and obstructed airways was <0.3 Pa, which is within the physiological limit needed to promote respiratory defense mechanisms. In summary, our model enables the study of airflow characteristics that may be impractical to assess experimentally.