272 EQUINE VETERINARY JOURNAL Equine vet. J. (2008) 40 (3) 272-279 doi: 10.2746/042516408X281216 Summary Reason for performing study: Computational fluid dynamics (CFD) models provide the means to evaluate airflow in the upper airways without requiring in vivo experiments. Hypothesis: The physiological conditions of a Thoroughbred racehorse’s upper airway during exercise could be simulated. Methods: Computed tomography scanned images of a 3-year- old intact male Thoroughbred racehorse cadaver were used to simulate in vivo geometry. Airway pressure traces from a live Thoroughbred horse, during exercise was used to set the boundary condition. Fluid-flow equations were solved for turbulent flow in the airway during inspiratory and expiratory phases. The wall pressure turbulent kinetic energy and velocity distributions were studied at different cross-sections along the airway. This provided insight into the general flow pattern and helped identify regions susceptible to dynamic collapse. Results: The airflow velocity and static tracheal pressure were comparable to data of horses exercising on a high-speed treadmill reported in recent literature. The cross-sectional area of the fully dilated rima glottidis was 7% greater than the trachea. During inspiration, the area of highest turbulence (i.e. kinetic energy) was in the larynx, the rostral aspect of the nasopharynx was subjected to the most negative wall pressure and the highest airflow velocity is more caudal on the ventral aspect of the nasopharynx (i.e. the soft palate). During exhalation, the area of highest turbulence was in the rostral and mid-nasopharynx, the maximum positive pressure was observed at the caudal aspect of the soft palate and the highest airflow velocity at the front of the nasopharynx. Conclusions and clinical relevance: In the equine upper airway collapsible area, the floor of the rostral aspect of the nasopharynx is subjected to the most significant collapsing pressure with high average turbulent kinetic during inhalation, which may lead to palatal instability and explain the high prevalence of dorsal displacement of the soft palate (DDSP) in racehorses. Maximal abduction of the arytenoid cartilage may not be needed for optimal performance, since the trachea cross-sectional area is 7% smaller than the rima glottidis. Introduction It is well known that the equine upper airway is subjected to marked variation in flow and pressure during both phases of respiration at maximal exercise (Derksen et al. 1986; Shappell et al. 1988; Nielan et al. 1992; Ducharme et al. 1994). Indeed, tracheal pressure fluctuations from -4905–2746.8 Pa and airflow velocity of 64–79 l/s during inspiration and expiration have been reported in horses during strenuous treadmill exercise at maximal exercise intensity (Ducharme et al. 1994; Lekeux and Art 1994; Tetens et al. 1996; Curtis et al. 2005, 2006). Left laryngeal hemiplegia and dorsal displacement of the soft palate are the most common causes of upper airway obstruction, but palatal instability, nasopharyngeal collapse, vocal cord collapse, medial collapse of aryepiglottic folds and epiglottic entrapment are also apparent (Dart et al. 2001; Parente et al. 2002; Lane et al. 2006a,b). The current study aimed to increase understanding of the mechanical implications of various upper airway obstructions beyond the measurements of airway pressures and flows and multiple derivations from those measurements (Derksen et al. 1986; Shappell et al. 1988; Lumsden et al. 1994; Tetens et al. 1996; Radcliffe et al. 2006). As an alternative to animal experiments, a physical model was created (Bayly and Slocombe 1997) of a horse’s upper airway from the external nares to the main stem bronchi by first filling the airway of a cadaver with plaster and then coating it with fibreglass. Air was then pumped through the nares and a tidal volume of 12 l reproduced at a frequency of 106 breaths/min with Reynolds number (a dimensionless number that is the ratio of inertial force to the viscous force in a fluid; Bird et al. 2007) as 63,000. As an alternative to human experiments, computational fluid dynamics (CFD) has been used extensively to model airflow through the human nasal cavity (Elad et al. 1993; Martonen et al. 2002; Allen et al. 2004; Lindemann et al. 2005; Xu et al. 2006); however, there have not been similar studies using horses to date. Both laminar and turbulent flow models using geometry reconstructed from scanned images have been used to study airflow in the human airway, for example: Martonen et al. (2002) and Allen et al. (2004) modelled paediatric upper airways, Xu et al. (2006) studied obstructive sleep apnoea in children, and Lindemann et al. (2005) looked at airflow after radical sinus surgery. Development of equine upper airway fluid mechanics model for Thoroughbred racehorses V. RAKESH, N. G. DUCHARME* † , A. K. DATTA, J. CHEETHAM † and A. P. PEASE ‡ Department of Biological and Environmental Engineering, College of Agriculture and Life Sciences, and † Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, New York 14853; and ‡ Department of Molecular Biomedical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina 27606, USA. Keywords: horse; flow modelling; upper airways; arytenoid abduction Video to demonstrate techniques used in this article will be available soon on our website, www.evj.co.uk *Author to whom correspondence should be addressed. [Paper received for publication 04.09.07; Accepted 20.12.07]