Computed Tomography to Assess the Brachycephalic Airway

Scansen BA.

in Conference Proceedings. American College of Veterinary Internal Medicine 2016.


The brachycephalic airway syndrome (BAS) is a disease of the nasal cavity, nasopharynx, and larynx related to early ankylosis of the basicranial epiphyseal cartilage. As a result of this early fusion, the longitudinal axis of the skull is shortened, particularly the upper jaw and upper respiratory tract. With a shortened nasal cavity, the nares and nasal cavity become occluded by growth of tissue and nasal conchae that have nowhere else to develop resulting in intranasal airflow obstruction. The soft palate extends caudal to the epiglottis due to the shortened upper jaw and results in stertorous breathing. Chronic increases in negative pressure within airway results in pharyngeal edema and laryngeal inversion. The disease occurs in the popular brachycephalic breeds including Pugs, English & French Bulldogs, Boston Terriers, Pekingese, Shih Tzu, etc. It is important to remember that BAS also occurs in feline breeds – in particular the Persian. In cats, the BAS has also been shown to impair nasolacrimal drainage resulting in ocular disease.1

Traditional evaluation of dogs with BAS involved direct visualization of the upper airway – nares, soft palate, and larynx. However, evaluation of the canine brachycephalic airway syndrome (BAS) by direct visualization is subjective and operator dependent. Not surprisingly, current surgical therapies show variable success. Recently, computed tomographic techniques to evaluate BAS in dogs have been published.2-5 These reports have involved dogs under general anesthesia and either intubated or with the mouth fixed open. This lecture will review CT findings in dogs with BAS and present an approach for CT evaluation of dogs with BAS under conscious sedation, which avoids distortion of the complex 3D morphology of BAS. Furthermore, advanced techniques to quantitate airflow resistance will be introduced, which allows for objective assessment of medical therapy or surgery on BAS severity.


Historical classification systems grouped BAS lesions into primary and secondary components. The primary components being stenotic nares, elongated soft palate, deviated nasal septum, nasopharyngeal turbinates, macroglossia, and redundant pharyngeal folds; the secondary components being everted laryngeal saccules, laryngeal collapse, tonsillar eversion, and mucosal edema. In a review of 90 dogs with BAS, 94% had an elongated soft palate, 77% had stenotic nares, 66% had eversion of the laryngeal saccules, and 56% had tonsillar eversion.6 In another study, stenotic nares were present in 43% of dogs, elongated soft palate in 86%, and laryngeal collapse in 53% of dogs with BAS.7

While the above abnormalities are typically considered the primary components of BAS, work by Gerhard Oechtering suggests that the traditional scheme is superficial as the areas addressed by conventional surgery are before and after the real problem – too small a nasal cavity resulting in intranasal obstruction.5 Similarly, mucosal contact points within the nasal cavity reduce the lumen for air passage and result in further intranasal obstruction.8 Complete evaluation of the degree of intranasal obstruction requires computed tomography or rhinoscopy. A report evaluated functional assessment of the degree of intranasal airway resistance with impulse oscillometry9, though this is likely only applicable in the research setting and has not been adopted in clinical cases. Treating the nares rostral to the nasal cavity by alarplasty and the palate caudal to the nasal cavity by staphylectomy may not be addressing the primary site of the problem that is obstruction within the nasal cavity. Laser-assisted turbinectomy has been described to address intranasal obstruction with six-month nasal patency reported at 84%.10

Tracheal hypoplasia is commonly considered in the context of BAS, though is primarily a condition of the English Bulldog. More recent work suggests that the tracheal diameter of Bulldog puppies may grow/improve with age, while this increase was not seen in puppies of other breeds.11 In addition to tracheal abnormalities in BAS, bronchial collapse also appears to be a co-finding in dogs with BAS.12 In a study of 40 dogs with BAS, fixed bronchial collapse was noted in 35 of the dogs and the degree of laryngeal collapse was positively correlated with severity of the bronchial collapse.12 While elongation of the soft palate is commonly considered in the evaluation of BAS, the presence of palate thickening has also been shown to occur in these dogs.2 Whether the thickening is primary or secondary to mucosal edema and swelling from the high intranasal pressures is unknown.

The presence of nasopharyngeal turbinates was originally reported in ~ 20% of dogs and cats with BAS as observed by retroflex endoscopy13, though recent reports suggest two-thirds of dogs with BAS have caudal aberrant turbinates.5 As noted above, this relates to expansion of normal structures into atypical locations due to the smaller nasal cavity.

The association of nasopharyngeal turbinates and clinical disease is not proven, but they likely play a role in airway obstruction and the clinical syndrome. Intranasal epidermoid cysts have also been reported in dogs with BAS, exacerbating upper airway obstruction.14 Last, gastrointestinal abnormalities appear commonly in dogs with BAS. A study of 73 dogs with BAS confirmed the presence of significant GI pathology in 74% of animals.15 Findings included esophageal and gastric inflammation, gastro-esophageal reflux, gastric stasis, and pyloric stenosis.15 Another study evaluating 50 dogs with BAS reported gastrointestinal signs in 88%.16 Therapy for GI disorders, together with surgical amelioration of BAS, has been shown to improve both respiratory and GI symptoms in dogs with BAS.17


Several reports have described the use of CT for evaluation of BAS in dogs. A report in 2007 of 23 dogs with BAS described two forms of aberrant conchae – rostral conchae obstructing the nasal passage and caudal conchae obstructing the choanae.18 In a subsequent series2, 26 dogs were evaluated with differing severity of respiratory symptoms compared to 5 control dogs. Dogs with severe BAS had thicker soft palates as compared to mildly affected dogs and controls; however, no differences were found between groups for soft palate length.2 Minimal cross-sectional area of the nasopharynx was slightly larger for less affected dogs, though significant differences between groups for normalized cross-sectional area were not found.2 A more recent study used CT to compare nasopharyngeal dimensions between Pugs and French Bulldogs.4 In this comparison, Pugs had a smaller cross-sectional area within the nasopharynx compared to the French Bulldogs; additionally, the soft palates of the Pugs were shorter and thinner than the French Bulldogs.4

The largest series of CT findings in dogs with BAS evaluated 132 dogs comprised of 66 Pugs, 55 French Bulldogs, and 11 English Bulldogs.5 Notable findings included abnormal nasal conchae in all dogs, resulting in variable degrees of intranasal obstruction; interconchal and intraconchal mucosal contact points were apparent in 92% of dogs.5 These same authors reported clinical improvement in these dogs when laser-assisted turbinectomy was performed to remove the aberrant conchae and open the nasal cavity.10

Data will be presented of CT findings in 23 dogs with BAS performed under conscious sedation without endotracheal intubation and with the mouth closed, representing non-distorted airway passages. Additionally, data will be presented from CT findings before and after conventional surgery for BAS in six dogs. Last, the effect of 0.05% oxymetazoline (Afrin) on CT findings will be shown, indicating that atomization of this decongestant into the nasal cavity of BAS is both tolerated and reduces mucosal swelling and contact points (Figure 1).


Current evaluation of dogs with BAS relies on history (presence of clinical signs) and physical examination (visual inspection of stenotic nares, soft palate, and larynx). In light of the spectrum of findings noted above and differing surgical approaches to this disease, it seems plausible to pursue a more quantitative assessment of the disease to

Figure 1

Axial images of the rostral nasal cavity (A, B) and at the level of the choanae (C, D) of an English bulldog with BAS before (A, C) and after (B, D) administration of oxymetazoline (Afrin). Images were acquired at the identical site before and 15 minutes after the decongestant; note the reduction in soft tissue opacity lining the nasal turbinates and choanae, reflecting reduced mucosal swelling and improved cross-sectional dimension of the air passages.

determine both the need for intervention as well as to measure the success of a given intervention. A few approaches have been reported to accomplish this quantitative assessment – impulse oscillometry9 and whole body barometric

plethysmography19. Both techniques require advanced equipment and impulse oscillometry requires the patient to be anesthetized. Whole body plethysmography does not yield a direct or linear measure of airway resistance; it has been reported to derive a predictive index of BAS severity based on several respiratory parameters derived from breathing curves. However, whole body plethysmography has the advantage of being measurable on the awake, non-sedated dog.

The author will present data on a novel approach to quantitation of airway resistance in dogs with BAS: computational fluid dynamic modeling of the nasal passage. Computational fluid dynamics (CFD) is a computer-based methodology that uses an imaging dataset to construct an anatomically accurate geometry of the upper respiratory tract, which can be extrapolated to a three-dimensional meshwork and allow mathematical simulation of airflow within this geometry (Figure 2). We are then able to use these CFD models to estimate airflow resistance within the airway under differing conditions.

Computed tomography datasets from brachycephalic dogs were used to investigate the feasibility of three- dimensional reconstruction and CFD modeling of airflow resistance. The transverse, bone algorithm datasets were imported into the imaging software for evaluation and generation of a volume meshwork. Semi-automated segmentation of the nasal passageway from the nares to the nasopharynx at the level of the caudal tip of the soft palate was achieved. Volumes of the airway were constructed into a three-dimensional meshwork (Figure 2). The volumetric dataset was then saved as a stereolithography computer aided design file and imported into the modeling software. The models were evaluated in a CFD software package and airflow simulated by applying a pressure gradient from the nasal passage to the pharynx. The CFD simulations demonstrated a significant increase in airflow and a significant decrease in overall airflow resistance after administration of the decongestant oxymetazoline (Afrin); analysis of the before and after surgery datasets is ongoing.


Generations of selective breeding have led to brachycephalic dog breeds characterized by a shortened craniofacial conformation and the clinical syndrome of BAS. Cross-sectional imaging such as computed tomography has enabled better depiction of complex anatomy that is grossly inaccessible due to physical barriers. Computed tomography and magnetic resonance imaging are the standard of care for evaluation of intranasal obstruction in humans. Similarly, CT evaluation of BAS can provide additional insight into the morphology and severity of the obstructive process in dogs. We have utilized CFD modeling derived from CT scans of dogs with BAS to construct an anatomically accurate geometry of the upper respiratory tract via generation of a three-dimensional meshwork. Further work will assess the utility of this method to quantify airflow resistance in affected dogs and determine optimal therapeutic strategies.

Figure 2

Airflow modeling from an English Bulldog. The CT data is manipulated to define the air-filled spaces of the nasal passage (A). The air passages are then extracted and modeled as a 3-dimensional surface geometry (B). Computer modeling of the air passage geometry is then performed and resistance calculated along the nasal passage with gradations in color signifying a change in resistance (C). In this dog, the greatest step-up in resistance occurs in the nasal turbinates (to the left of the image).


1. Schlueter C, et al. J Feline Med Surg. 2009;11:891–900.

2. Grand JG, et al. J Small Anim Pract. 2011;52:232–239.

3. Kaye BM, et al. Vet Radiol Ultrasound. 2015;56:609–616.

4. Heidenreich D, et al. Vet Surg. 2016;45:83–90.

5. Oechtering GU, et al. Vet Surg. 2016;45:165–172.

6. Fasanella FJ, et al. J Am Vet Med Assoc. 2010;237:1048–1051.

7. Torrez CV, et al. J Small Anim Pract. 2006;47:150–154.

8. Schuenemann R, et al. J Am Anim Hosp Assoc. 2014;50:237–246.