Mechanics of airflow in the human nasal airways

Abstract

The mechanics of airflow in the human nasal airways is reviewed, drawing on the findings of experimental and computational model studies. Modelling inevitably requires simplifications and assumptions, particularly given the complexity of the nasal airways. The processes entailed in modelling the nasal airways (from defining the model, to its production and, finally, validating the results) is critically examined, both for physical models and for computational simulations. Uncertainty still surrounds the appropriateness of the various assumptions made in modelling, particularly with regard to the nature of flow. New results are presented in which high-speed particle image velocimetry (PIV) and direct numerical simulation are applied to investigate the development of flow instability in the nasal cavity. These illustrate some of the improved capabilities afforded by technological developments for future model studies. The need for further improvements in characterising airway geometry and flow together with promising new methods are briefly discussed.

Introduction

In all mammals, the upper airway is the portal to the respiratory system and all respired air passes through it on both inspiration and expiration. The upper airway is a complicated structure comprising the mouth and nasal passages placed in parallel, with the nose itself containing two parallel pathways for airflow. Virtually all mammals are obligatory nose breathers; however, at other than relatively low respiratory flow rates, humans and other primates usually breathe preferentially via the mouth (Proctor, 1986). Ventilation is not only cyclical within the upper airway, but usually unequally divided between the two nasal cavities, with the bias of flow alternating between sides over a period of time – the nasal cycle (Eccles, 2000) – that is orders of magnitude larger than a single respiratory cycle. Within the passages of the airway, flow is thus time dependent on both long and short time scales.

The gross architecture of the nasal cavities varies considerably between species (Negus, 1958) for many possible reasons, but particularly to accommodate differing head and jaw shapes associated with eating demands. Consequently, the internal airflow conditions vary considerably between species and in order to understand the details of the airflow and transport properties in the human nose it is important to focus studies explicitly on the human.

The principal physiological function of the nose is to heat and humidify inhaled air (see review in this issue by Elad et al., 2008) and remove noxious materials from the air stream, protecting the delicate distal pulmonary structures. The nose also contains the olfactory apparatus to enable smelling of substances for beneficial or defence purposes (Proctor, 1977). Summary descriptions of the anatomy, physiology and function of the human nasal airways may be found in many works, see for example Mygind and Dahl (1998). The comprehensive text of Lang (1989) includes numerous images of dissections of the airways, compilations of the dimensions of notable anatomical features, and descriptions of variations in morphology.

The technical challenges posed by the task of obtaining detailed, spatially resolved, in vivo measurements of the conditions within the nasal airways have yet to be overcome. This is due, not only, to the tortuous geometry of the passages, their narrowness and biological responsiveness to touch, but because any inserted probe is likely to cause perturbations in the flow and introduce experimental artefact. In vivo measurement of functional parameters, such as pressure drop, humidification or noxious gas uptake, is essentially restricted to global information based on measurements up- and down-stream of the airway. Internal details are interpreted from this data; however, the complexity of the conduit morphology means that such implied distributed information is based on surmise.

To improve our understanding of nasal function, it is necessary to undertake model studies in which experimental exploration of distributed function can be made explicitly. However, modelling introduces simplifications which may lead to uncertainty or inaccuracy in translation of the findings back to real life. Both physiologists and modellers need to be fully aware of the shortcomings of modelling in order to make properly informed interpretations of the information obtained in such studies.

The aim of this review is to provide a critical evaluation of issues involved in modelling airflow in the human nose. The nature and quality of experimental data obtained via model studies are considered, and the potential of newer techniques to improve specific aspects of modelling are discussed. Rather than attempting an exhaustive literature critique, references are made frequently to recently published papers containing useful, broad, critical reviews of a particular topic, or which provide an account of the application of particular techniques, that may be used to explore the field further.