Inside the Brachycephalic Nose: Intranasal Mucosal Contact Points

Introduction

Reported features of brachycephalic syndrome (BS) include an elongated and thickened soft palate, stenotic nares, laryngeal collapse, everted saccules, and sometimes, a hypoplastic trachea.^1–5 The congenitally shortened skull bones of brachycephalic dogs lead to a relative hypertrophy of soft tissues within the head.⁵ All that leads to narrowing of the upper airway, which increases airway resistance and in the consequence requires greater inspiratory effort.³ That inspiratory effort leads to edema and inflammation caused by chronic barotrauma during respiration, causing further airway narrowing and leading to a vicious cycle of respiratory problems.³ Several breeds are affected by BS in varying degrees. Boston terriers, Cavalier King Charles spaniels, and boxers seem to be only mildly affected; however, respiratory problems associated with brachycephaly has become an increasing problem in pugs and French and English bulldogs in recent years, which is likely ascribed to breeding for extreme brachycephaly. Other breeds affected by BS include Pekingese, shih tzu, and Lhasa apsos.⁵

The soft palate, nares, and saccules have been surgically addressed for decades already, but no new approaches have been described recently, although clinical success of traditional surgical methods varies.^3,5 BS is a multimodal disease with airway stenoses in more than one location. If not all locations are addressed concurrently, one cannot expect a clinically healthy animal after surgery if certain stenoses remain. It is hard to differentiate what locations contribute the most to clinical symptoms in a certain dog, and those locations vary between breeds and individuals. Thorough airway examination by means of computed tomography and endoscopy should always be performed before any stenoses are surgically addressed to assess all relevant locations.

It is surprising that intranasal problems have been overlooked for a long time, even though the shortened nose is the most eye-catching feature of brachycephalic dogs. In 2007, Oechtering et al. first reported on aberrant turbinate growth as another feature of brachycephalic dogs.⁶ Ginn et al. (2008) found pugs to be the breed with the highest prevalence of nasopharyngeal turbinates.⁷ Such turbinates can, similarly to nasopharyngeal polyps, block the nasopharyngeal lumen significantly and contribute to further upper airway obstruction. The turbinates of brachycephalic dogs also differ histologically from those of normocephalic dogs.⁸ Lippert et al. (2010) proved by impulse oscillometric measurements that the intranasal resistance is elevated in many brachycephalic dogs.⁹ Nasal airway resistance provides half the airflow resistance of the entire system of respiratory air passages in humans; therefore, for dogs, dependent on nasal breathing for their thermoregulation, this may be of even greater importance.^10,11

During rhinoscopic examination of brachycephalic dogs, the authors observed a high prevalence of intranasal mucosal contact (of varying degrees) with surrounding structures. The study authors hypothesized that such contact is a common entity in brachycephalic dogs, but rare in normocephalic dogs. To investigate the prevalence of mucosal contact and to classify the location of mucosal contact points in brachycephalic dogs, this prospective controlled prevalence study was conducted.

Materials and Methods

Cases that were referred for surgical treatment of BS between January 2009 and February 2011 were evaluated by rhinoscopy for conchal contact. Eighty-two brachycephalic dogs were evaluated, 42 of which were pugs and 40 were French bulldogs. The control group was comprised of 25 normocephalic dogs that received endoscopy of the respiratory tract for reasons other than nasal pathology. Reasons for endoscopy in those dogs included stick injury, tracheal collapse, laryngeal paralysis, and respiratory tract diagnostics due to unspecific respiratory symptoms. Owners were informed about rhinoscopy being part of the authors’ standard respiratory tract endoscopy protocol. Dogs included in the control group were Labrador retrievers (n = 4), Yorkshire terriers (n = 3), dachshunds (n = 3), mixed-breed (n = 2), and one each of the following: German shepherd dog, Weimaraner, Bernese mountain dog, Bolonka swetna, flat-coated retriever, Irish wolfhound, Russian terrier, leonberger, Kromfohrlander, Jack Russell terrier, German longhaired pointer, golden retriever, and West Highland white terrier. Normocephalic dogs were divided into two groups based on size (small normocephalic dogs ≤ 11 kg and large normocephalic dogs > 11 kg).

Endoscopy was performed in sternal recumbency under general anesthesia. Dogs were premedicated with levomethadone^a (0.1–0.2 mg/kg IV) and diazepam^b (0.1 mg/kg IV), induced with either propofol^c (1–4 mg/kg IV) or a combination of ketamine^d and xylazine^e (1–2 mg/kg IV and.1–1 mg/kg IV, respectively), and maintained with isoflurane^f in an O₂/air mixture.

Results

In the brachycephalic group, 53 dogs were male (64.6%), 2 were castrated males (2.4%), 16 were females (19.5%), and 11 were spayed females (13.4%). Mean body weight ± standard deviation in pugs was 8.9 kg ± 1.8 kg (range, 4.4–11.9 kg), and in French bulldogs it was 11.4 kg ± 2.3 kg (range, 6.6–17 kg). Mean age at the time of examination was 35 ± 14.5 mo in pugs (range, 5–69 mo) and 29.9 mo ± 20.3 mo in French bulldogs (range, 8–81 mo).

In the normocephalic control group, 9 dogs were intact males (36%), 2 were neutered males (8%), 12 dogs were intact females (48%), and 2 were spayed females (8%). Weights ranged from 3.2 kg to 70 kg (mean, 26.4 kg ± 20.4 kg). Ages ranged from 12 mo to 166 mo (mean, 82 mo ± 43.7 mo). Fifteen dogs weighed > 11kg and 10 dogs weighed ≤ 11kg.

Septal deviations were present in 40 pugs (95.2%), which were equally distributed between the right and left side. In addition, 11 French bulldogs (27.5%), also had septal deviations, 7 of which had deviations to the right and 4 to the left. In the normocephalic group, a septal deviation was present in four of the dogs (16%), which were equally distributed between the left and right side. Three of those dogs belonged to the group weighing < 11 kg (Table 1).

TABLE 1 Distribution of Septal Deviations per Dog*

*Data are presented as number (percent).

†

P  =  .000 (comparison between pugs and French bulldogs)

§

P  =  .62 (comparison between French bulldogs and small normocephalic dogs)

**P  =  .031 (comparison between small and large normocephalic dogs)

View Fullscreen

The plica recta was enlarged in many brachycephalic dogs and consequently contacted the septum, the plica alaris, or both. In pugs, the plica recta was usually in contact with the either plica alaris or both the plica alaris and the septum. In French bulldogs, contact with the septum was much more common. In the group of normocephalic dogs ≤ 11 kg, 11 of the 20 nasal cavities (55%) had a hypertrophic plica recta. In five cases, the plica recta contacted the septum, contacted the plica alaris in five cases, and contacted both in one case. In the group of dogs weighing > 11 kg, the plica recta was in contact with the septum in only 3 of 30 nasal cavities (10%). That difference was statistically significant as summarized in Table 2.

TABLE 2 Distribution of Contact of the Plica Recta per Nasal Cavity*

*Data are presented as number (percent).

†

P  = .000 (comparison between pugs and French bulldogs)

§

P  = .014 (comparison between French bulldogs and small normocephalic dogs)

**P  = .001 (comparison between small and large normocephalic dogs)

View Fullscreen

There were only three pugs and one French bulldog (7.1% and 2.5% within their breed) in which there was no mucosal contact in either nasal cavity. Thirteen pugs and one French bulldog had one nasal cavity without mucosal contact.

Two nasal cavities could not be evaluated (one in a pug and one in a French bulldog). Of the 162 nasal cavities that could be evaluated, 140 had at least one mucosal contact point (87%), and only 22 showed no mucosal contact at any location (13.6%). In the French bulldogs, 76 of 79 nasal cavities examined (96%) had mucosal contact. In pugs, there was mucosal contact in 64 cavities (77%) and no contact in 19 cavities (13%). The difference between French bulldogs and pugs in terms of contact was statistically significant (P <  .0005). See Tables 3 and 4 for a summary of the different locations and their statistical differences. French bulldogs had three mucosal contact points on average, whereas pugs only had 1.7. That difference was also statistically significant (P <  .0005) as shown in Table 5 and Figure 1.

TABLE 3 Distribution of the Intranasal Mucosal Contact in Different Anatomic Locations per Nasal Cavity*

*Data are presented as number (percent).

†

One nasal cavity of each breed could not be evaluated due to a foreign body.

View Fullscreen

TABLE 4 Distribution of Internasal Mucosal Contact in Brachycephalic Dogs Only*

*Data are presented as number (percent).

View Fullscreen

TABLE 5 Average Number of Contact Points per Nasal Cavity

*P  =  .000 (comparison between pugs and French bulldogs)

†

P  =  .152 (comparison between small and large normocephalic dogs)

Note: P  =  .000 for a comparison between the brachycephalic and the normocephalic group. SD, standard deviation.

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

In the normocephalic dogs, contact between intranasal mucosal structures was extremely rare. When it occurred, it was usually observed in a single location rather than throughout the nasal cavity. In 19 dogs (76%) there was no mucosal contact between intranasal structures. Contact at a single location was evident in five dogs. In one dog, a 3.2 kg Yorkshire terrier, contact was identified in more than one location. Contact occurred in both cavities between the CNV and septum and in one cavity the CNV lamellae were in contact with each other as well as with the plica basalis.

Of the 50 evaluated nasal cavities, mucosal contact could be detected in 7 cavities (14%). Only one cavity had interlamellar contact of the branches of the CNV, and that contact occurred only in a single location. In six nasal cavities (12.5%), contact between the CNV and the septum was present, and two had a septal deviation to the side of contact. In one case, contact of CNV lamellae with the plica basalis was present, and in another case, contact was with the plica recta. The differences in intranasal mucosal contact of the normocephalic group were statistically significant compared with the brachycephalic group, but not between small and large normocephalic dogs (Table 3, Figures 2–5). Small normocephalic dogs had a contact point average of 0.3 and large normocephalic dogs had a contact point average of 0.1. That difference was not significant (Table 5, Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Gender did not have a significant influence on either the presence or absence of mucosal contact. Weight showed no correlation with mucosal contact.

Discussion

This study shows that mucosal contact between intranasal structures is significantly more common in brachycephalic dogs than normocephalic dogs of all sizes. Changes in the plica recta seem to be different, as it is often enlarged in relation to the size of the nasal cavity in miniature dogs with noses of normal length ratio. Reasons for that finding could not be identified.

It is likely that the mucosal contact described in this study is responsible for the intranasal airway obstruction from which many brachycephalic dogs suffer. To the authors’ knowledge, this mucosal contact is a previously unreported contributing factor to the respiratory problems of brachycephalic dogs. Lippert et al. (2010) conducted a study measuring intranasal airway resistance via impulse oscillometry and proved that brachycephalic dogs suffer from elevated resistance.⁹ Impulse-oscillometric rhinomanometry is the objective gold standard in humans to determine whether intranasal obstruction is present or not.¹² In dogs, rhinomanometry is difficult to perform, requires a lot of practice, and is time consuming. Because rhinomanometry cannot be performed in conscious dogs, it contributes to significantly longer anesthetic times; therefore, it cannot be considered a routine diagnostic tool. Assessing intranasal obstruction endoscopically is somewhat subjective, but it takes minimal time to evaluate mucosal contact points, which can serve as an indicator of probable intranasal obstruction. In people, it has been shown that rhinoscopic assessment of nasal obstruction is closely correlated with objective measurements of nasal patency.¹³ It can, therefore, be recommended as a routine instrument for assessing the necessity of intranasal surgery in brachycephalic dogs. A combination of both, rhinoscopic assessment and rhinomanometry would be the ideal assessment method. Future studies that compare endoscopic evaluation and rhinomanometric measurements would be interesting.

The difference in mucosal contact seen in French bulldogs and pugs was statistically significant. Only 30% of pugs showed interlamellar contact as opposed to 79% of French bulldogs. Interlamellar mucosal contact of the CNV is the main contributor to blockage of nasal airflow. The lamellae of the CNV take up the most space inside the nasal cavity, so the more contact, the smaller the lumen of nasal passageways. The higher incidence of mucosal contact in French bulldogs may explain why turbinectomy is much more often warranted in that breed than in pugs. It is surprising that pugs, which have the shorter noses compared with French bulldogs, actually have less intranasal mucosal contact. French bulldogs tend to have thicker soft palates than pugs. It is possible that other mucosal structures, such as the nasal mucosa, are hypertrophic in those dogs as well.¹ Most pugs had a marked plica recta enlargement, often touching the septum and the plica alaris. That causes an obstruction already in the nasal vestibule, whereas pugs’ cavities often seem to be less obstructed than those of French bulldogs. In dogs with hypertrophic plicae rectae alone, removal of these might be sufficient to relieve intranasal obstruction.

If contact between nasal structures occurred in the normocephalic dogs, it was usually only in a single location. Dogs that had a septal deviation showed contact between lamellae and the septum on the convex side. Studies in people have shown that postural stimuli, such as lateral recumbency, causes vascular responses in the nasal mucosa and, therefore, temporary swelling.¹⁴ It cannot be excluded that some of the dogs may have been in lateral recumbency before the examination, which may have contributed to further mucosal contact. However, that would be true for both, brachy- and normocephalic dogs.

The reported prevalence of nasal mucosal contact points in humans ranges from 4% to 55% among rhinology patients.^15,16 Definition of contact points differs between studies. For example, Goldsmith et al. (1993) defines a contact point as one that is still present after decongestion, whereas Bieger-Farhan et al. (2004) includes all contact points, even if temporary, in their study.^16,17 It would be interesting for future studies to evaluate what contact points resolve after use of a decongestant like xylometazoline in brachycephalic dogs. Detumescence of hypertrophic nasal mucosa would likely lead to resolution of most, if not all, contact points. In cases in which the mucosal contact is caused by either underlying bone hypertrophy or marked septal deviation (not only hypertrophic mucosa), the contact would not resolve after decongestion. In human medicine, there is controversy as to whether such contact points cause facial pain in people.¹⁶ However, there seems to be no doubt that contact points cause nasal blockage as well as loss of smell.^16,18 Contact is believed to result from either septal deviations or from conchal hypertrophy, both of which are present in brachycephalic dogs.¹⁸

Conchae may be hypertrophic for physiologic, pathophysiologic, or anatomic reasons.¹⁹ A compensatory hypertrophy at the concave side of a septal deviation can be considered a physiologic response, which attempts to reduce turbulent airflow.²⁰ Human patients with rhinitis show a pathologic hypertrophy as a response to the inflammation.¹⁹ Anatomic conditions in people include bony hyperplasia of the inferior turbinate and concha bullosa of the middle turbinate.¹⁹ Hypertrophy of turbinates occurs at either the inferior or middle turbinate, with the inferior turbinate contributing much more to impaired nasal airflow if hypertrophic.²¹ It can be either caused by the soft tissue of the concha, the bony turbinate, or a combination of both. Soft-tissue hypertrophy is most common and is believed to develop in the setting of chronic rhinitis or other conditions that cause chronic mucosal inflammation. Mixed hypertrophy most likely involves anatomic bony hypertrophy in the setting of chronic rhinitis.²¹

In the brachycephalic dog, it is likely that a mixed hypertrophy is present. Histologic studies of conchae of brachycephalic dogs found thickened underlying bone covered by a thickened mucosa with enlarged vessels compared with samples from normocephalic dogs.⁹ In most cases, the mucosa showed signs of chronic rhinitis. It is not clear whether the bony hypertrophy is the result of a primary genetic condition in these dogs (with the mucosal hypertrophy developing secondary to chronic rhinitis) or whether both conditions are just another inherited disorder in brachycephalic dogs. In the former case, chronic rhinitis would have to occur at a very early age in those dogs because puppies as young as 5 mo of age can have hypertrophic conchae (in the authors’ experience). At present, the authors consider it more likely that both bony and mucosal hypertrophy of the turbinates, are another genetic problem in brachycephalic dogs, and chronic rhinitis may develop secondarily as a result of impaired nasal airflow, thereby reducing mucociliary clearance.¹⁸ However, compensatory hypertrophy of conchae that grow into concave sides of septal deviations are another likely entity, at least in pugs.

The reason for the anatomically deformed conchae in brachycephalic dogs is probably the shortened nasal cavity. In brachycephalics, there is postnatal growth inhibition of the splanchnocranium, and conchal growth in dogs is not terminated after birth.²² Because there is less space in the nasal cavities of brachycephalic dogs, conchae find different paths to grow and form so-called rostral and caudal aberrant turbinates.^6,23 Those conchae are too large in relation to nasal cavity size, contributing to mucosal contact with surrounding structures.

There may be intrinsic controls on conchal growth, but they also seem to adapt their growth behavior as needed. The dead space on the concave side of a septal deviation causes turbulence in the incoming airflow. To reduce that, the relevant conchae become hypertrophied.²⁰ Apparently, conchal cells receive information regarding airflow and resultant pressure and shear stresses. It is well known in cardiovascular fluid mechanics that cultured aortic endothelial cells exposed to a certain physiologic shear stress align in the direction of blood flow, whereas those exposed to lower shear stresses do not.²⁴ It was also shown that shear stresses influence the cytoskeletal appearance of nasal epithelial cells.²⁵ Thus, it is feasible that stenotic nares hinder the incoming air streams early in life, which leads to loss of physical stresses (shear stresses and pressure generated by the air streams) inside the nasal cavity. This might lead to undirected growth of conchal cells because of the missing information provided by those physical stresses. Possibly, physical stresses in those unaerated nasal cavities are only applied once contact with another structure occurs.

As to the question if mucosal contact is either acquired or congenital, the answer must be both. Tendency to develop mucosal contact is probably a genetic phenomenon given the anatomic circumstances of shortened nasal cavities and hypertrophic mucosal tissues, which is further contributed to by septal deviations. However, as conchae have not terminated growth at the time of birth, the contact points themselves only occur once conchae have grown.²² The study authors have performed CT scans from brachycephalic neonates that were born dead in a cesarean section and found only conchal buds that were small and not in contact with surrounding tissues. As the young dog grows, conchae as well as soft tissues grow, and problems associated with brachycephaly usually only begin at a later age, not right after birth. The soft palate, for example, is probably genetically predetermined too large in brachycephalic dogs, but due to increased inspiratory effort, edema and secondary enlargement is the consequence. Similarly, swelling of nasal mucosa because of inflammation further contributes to mucosal contact points in hypertrophic conchae.

Early correction of obstructing nares might be able to influence conchal growth and prevention of contact points by promoting better airflow and consequently increased shear stresses through the nasal cavity. Such surgery would be required prior to the termination of conchal growth. Some dogs involved in this current study were as young as 5 mo of age, suggesting that evaluation and treatment would be necessary even before that age. However, from an ethical point of view, it is unacceptable to tolerate breeding that requires regular surgery in adolescent animals to provide basic needs, like breathing. Instead, breeding programs in brachycephalic dogs should focus on animals with longer noses and wider nares, which may potentially lead to less obstructing conchal growth.

One limitation of this study is that the very caudal aspects of the nasal cavity could not be evaluated for mucosal contact. To visualize those areas, the endoscope has to be passed beyond the rostral CNV, which presses the concha laterally. That pressure might either create mucosal contact with other structures or push surfaces away from each other, hindering reliable evaluation. Furthermore, injuries to the conchae or mucosa might lead to bleeding, thus hindering visualization. The current study evaluated approximately the rostral one-half to two-thirds of the nasal cavity for mucosal contact, but it is quite possible that further contact was present in the more caudal areas of the nasal cavities.

Another limitation of this study is that clinical evaluation was not assessed. As BS is a multimodal disease with stenoses present in many different locations, it would be hard to judge clinical signs in correlation with intranasal contact points alone. Respiratory symptoms in brachycephalic dogs must always be seen as a consequence of a combination of anatomic deformities, and some deformities might contribute to the symptoms in one dog more than in another. The authors are currently conducting a study involving a systematic owner questionnaire that will hopefully answer some of the questions regarding the origin of clinical signs.

All dogs in this study were concurrently treated for stenotic nares and elongated soft palates. Laryngeal saccules were present and treated in a few of the dogs, but the saccules were not evaluated in the current study and will be the subject of future studies. Most dogs were not surgically treated before the examination; however, there were some dogs that had undergone surgery of their soft palate alio loco without clinical improvement before they were presented to the authors.

Brachycephalic dogs with mucosal contact may have been overrepresented in this study population because the dogs that were referred for treatment were suffering from severe brachycephaly and related clinical signs. Nevertheless, all brachycephalic dogs undergoing surgery should be evaluated for the presence of intranasal mucosal contact and subsequent intranasal obstruction. In animals where those are present, nares surgery alone would probably not be sufficient, and the intranasal obstruction should be removed as well. Those obstructions can be addressed by removal of turbinates, for example by a laser-assisted turbinectomy as described by Oechtering et al. (2007).²⁶ During the procedure, a diode laser fiber is introduced endoscopically into the nasal cavity to dissect the attachment of the turbinate from its origin. Thereafter, the turbinate can be removed with a small forceps under endoscopic guidance. Most often, only the CNV as the main contributor to nasal obstruction is removed. If the CNM is contributing to nasal obstruction as well, it can be removed partially by the same procedure.

As in other studies involving dogs with BS, male dogs were overrepresented in the brachycephalic group.^1,27 Septal deviations were significantly overrepresented in pugs compared with the other brachycephalic breeds. That may be due to the fact that pugs have the shortest nasal cavities of all brachycephalic breeds, leaving no space for a straight septum and forcing other intranasal structures to grow into any available space. However, septal deviations can occur in other dogs, too and seem to be present more often in small dogs. Septal deviations were present in 30% of the normocephalic dogs weighing ≤ 11 kg in this study. In pugs, septal deviations were always severe and bilateral, resembling a type 4 deviation according to the classification that was proposed by Mladina for human septal deviations, also known as an S-shaped deviation (Figures 6A, B).²⁸ If septal deviations were present in French bulldogs and normocephalic dogs, they were less severe and usually C-shaped, which could potentially be classified as a type 3 deviation.²⁸

View Figure

• Download as Powerpoint
• Download Figure

In humans, deviations of the septum are not uncommon and are reported in between 77% and 92% of people.²¹ It is believed that anterior deviations, the most common form in people, are primarily related to extrinsic forces, such as pressure during birth or other injuries, whereas posterior deviations have a genetic component.²⁹ Extrinsic forces can probably lead to mild septal deviations as are sometimes found in larger normocephalic dogs. Because almost all pugs have severe deviations, it is more likely a genetic condition in those dogs. Septal deviations can also contribute to mucosal contact and impaired nasal breathing.³⁰

Conclusion

Mucosal contact of intranasal structures was identified as another feature of brachycephaly that is only rarely found in normocephalic dogs. Mucosal contact points reduce the lumen of intranasal passageways and can cause intranasal obstruction in brachycephalic dogs. Septal deviations and hypertrophic conchae contribute to mucosal contact and impaired nasal breathing. The results of this study suggest that all dogs undergoing surgery for BS should be evaluated for intranasal contact points. Dogs with extensive mucosal contact might benefit from removal of obstructing conchae.