Radiographic tracheal dimensions in brachycephalic breeds before and after surgical treatment for brachycephalic airway syndrome

Abstract

The objective of this retrospective study was to compare preoperative and postoperative tracheal measurements in brachycephalic dogs undergoing surgical treatment of brachycephalic airway syndrome (BAS) using 2 radiographic methods (TD:TI and TT:3R). Medical records of 24 brachycephalic dogs with surgically corrected BAS were reviewed. Represented breeds included English bulldogs (n = 9), French bulldogs (n = 9), pugs (n = 5), and a Japanese chin (n = 1). Dogs with BAS that had postoperative thoracic radiographs ≤ 1 week after surgery showed no significant difference in the mean TD:TI ratios (P = 0.06) or mean TT:3R ratios (P = 0.19) when comparing preoperative to postoperative ratios. Dogs that had postoperative thoracic radiographs > 1 week after surgery showed no significant difference in the mean TD:TI ratios (P = 0.06) or mean TT:3R ratios (P = 0.32) when comparing preoperative to postoperative ratios. According to these findings, surgical correction for BAS does not resolve or improve tracheal hypoplasia in brachycephalic breeds.

Introduction

Brachycephalic airway syndrome (BAS) is a cause of respiratory distress in brachycephalic breeds. Breeds most commonly affected are English bulldogs, French bulldogs, pugs, and Boston terriers (1,2). This syndrome is comprised of a number of components. Primary components of BAS include the following: stenotic nares, elongated soft palate, hypoplastic trachea, redundant pharyngeal tissue, and possible aberrant nasal conchae (1,3,4). Secondary components include everted laryngeal saccules, pharyngeal edema, and laryngeal collapse (4). Clinical signs often reported by owners include heat stress, and exercise intolerance, snoring, inspiratory dyspnea, and occasionally cyanosis and syncopal episodes in severe cases.

Tracheal hypoplasia is a congenital trait (5–8) and is a concurrent finding in many patients with BAS (1,5,6,9,10). Affected dogs have small tracheal cartilage rings with overlapping ends and a narrow or absent dorsal tracheal membrane (6,7). Tracheal hypoplasia typically involves the whole trachea, with a uniformly narrowed lumen from the larynx to the carina. English bulldogs are more commonly affected by tracheal hypoplasia than are other breeds (6–8). In the absence of concurrent pulmonary disease, tracheal hypoplasia is usually not associated with clinical signs (5–7). Tracheal hypoplasia may exacerbate the respiratory signs associated with BAS because of increased resistance to airflow, but it was not associated with worse outcomes after surgical correction of BAS (1,10,11). Tracheal hypoplasia has been considered a negative prognostic indicator when it occurs in association with BAS or bronchopneumonia (5).

Palpation, radiography, tracheobronchoscopy, and fluoroscopy have been used to diagnose tracheal hypoplasia (5–8,11,12). Two radiographic methods have been described to evaluate the dimensions of the tracheal diameter in dogs and to distinguish between hypoplastic and non-hypoplastic tracheas: the tracheal lumen diameter to thoracic inlet distance ratio (TD:TI) (11) and the ratio between the thoracic tracheal luminal diameter and the width of the proximal third of the third rib (TT:3R) (13). Although neither method has been shown to be superior (5), the most commonly used method is TD:TI with the benefits including that it is independent of the size of the patient and phase of respiration (11,12). There is no information in the literature about TD:TI or TT:3R tracheal measurements in dogs before and after surgical treatment of BAS.

Tracheal diameters are not necessarily static and were shown to improve in 6 bulldog puppies after resolution of bronchopneumonia (14). Another study showed a change in the tracheal diameter during phases of respiration in healthy dogs as detected via computed tomography (CT) (15). There is speculation that the size of the trachea may change with age and patient growth; however, no studies proving this are available in the literature. Although tracheal hypoplasia is considered a congenital trait, there has been no prior investigation on whether or not tracheal dimensions would increase by decreasing airway resistance and improving airflow with surgical correction of BAS.

The purpose of this study was to compare preoperative and postoperative tracheal measurements in brachycephalic dogs undergoing surgical treatment of BAS using 2 radiographic methods (TD:TI and TT:3R) and to evaluate for changes in tracheal dimensions and possible resolution of tracheal hypoplasia. We hypothesized that the tracheal dimensions would not change either ≤ 1 week or > 1 week after brachycephalic airway surgery using either the TD:TI or TT:3R method.

Materials and methods

Medical records from an academic, referral veterinary hospital (University of Florida) were searched to identify brachycephalic dogs that received surgical treatment for BAS between July 1, 2008 and July 1, 2018. Brachycephalic breeds with surgically corrected brachycephalic airway syndrome and both preoperative and postoperative thoracic radiographs were retrospectively collected.

Surgical treatment may have included any one of the following treatments or a combination of the following: stenotic nares resection, staphylectomy, and surgical excision of everted saccules. Right and left lateral recumbent radiographs from 24 brachycephalic dogs were evaluated by a single Board-certified radiologist using the 2 different ratios.

Three-view thoracic radiographs were taken before and after surgery for each dog. TD:TI and TT:3R ratios were calculated and correlated on preoperative and postoperative (Figure 1) right and left lateral thoracic radiographs. When using the TD:TI ratio for tracheal dimensions, the thoracic inlet (TI) was measured from the first thoracic vertebra to the manubrium, and the tracheal diameter (TD) was measured where the TI line intersects the trachea (11). The TD:TI ratio for hypoplastic trachea was defined as < 0.16 in non-bulldog brachycephalic dogs and < 0.12 in English bulldogs. The TT:3R method used the ratio between the thoracic tracheal luminal diameter and the width of the proximal third of the third rib, and a TT:3R ratio of < 2.0 was used to define tracheal hypoplasia (6–8).

Figure 1.

Preoperative (A) and postoperative (B) right lateral thoracic radiographs demonstrating TD:TI (tracheal lumen diameter and thoracic inlet distance) ratio measurements on a single English bulldog patient. Preoperative (C) and postoperative (D) right lateral thoracic radiographs demonstrating TT:3R (thoracic tracheal luminal diameter and the width of the proximal third of the third rib) ratio measurements on the same English bulldog patient.

Each radiographic method was used both before and after surgery to determine the number of hypoplastic and non-hypoplastic dogs. Preoperative and postoperative thoracic radiographs were evaluated in both left and right lateral positions and compared using both radiographic methods. The cases were further subdivided into 2 groups. Group A (n = 12) consisted of cases in which thoracic radiographs were taken before surgery, and postoperative radiographs were taken ≤ 1 wk after surgery. Group B (n = 15) consisted of cases in which thoracic radiographs were taken before surgery, and postoperative radiographs were taken > 1 wk after surgery. Three patients were included in both Groups A and B, as they had postoperative radiographs taken ≤ 1 wk and > 1 wk after surgery.

Statistical analysis

The following fixed effect linear model was used to analyze each response variable using the SAS procedure MIXED (SAS/ STAT 14.1; SAS Institute, Cary, North Carolina, USA):

$$\text{Response}=\text{Orientation}+\text{Group}+\text{Orientation}\times \text{Group}+\text{Time\hspace{0.17em}}(\text{Group})+\text{Orientation}*\text{Time\hspace{0.17em}}(\text{Group}),$$

where: Orientation (L, R) refers to the aspect from which the radiographs were taken after surgery, Group (A, B) was explained in the last paragraph, and Time (Pre, Post) refers to the time the radiographs were taken. The nested approach for the time factor was chosen because the post time was the basis for the group assignment. Normality of residuals was evaluated graphically as recommended by Kozak and Piepho (16). Means of interest and associated standard errors were calculated using the LSMEANS statement of the above-named procedure.

Results

Twenty-four brachycephalic dogs were identified that met the study criteria.

Signalment

Females (n = 13) and males (n = 11) were diagnosed with and treated for BAS during the study period. Represented breeds included English bulldogs (n = 9; 37.5%), French bulldogs (n = 9; 37.5%), pugs (n = 5; 20.8%), and a Japanese chin (n = 1; 4.2%). Median age at presentation was 38 mo (mean: 41.4 ± 30.7 mo, range: 3 to 110 mo; N = 24).

Surgical treatment and findings

Surgical data were available for all 24 dogs in this study. A total of 23/24 (95.8%) staphylectomies or folded flap palatoplasties were performed. Nares resection was performed in 20/24 (83.3%) cases, and laryngeal saccules were excised in 19/24 (79.1%) cases. The most common surgical treatment performed for BAS was a combination of a nares resection, staphylectomy/palatoplasty, and laryngeal saccule excision in 13/24 cases (54.1%). A combination of nares resection, staphylectomy/palatoplasty, and laryngeal saccule excision with a temporary tracheostomy was performed in an additional 2/24 cases (8.3%), and 1/24 cases (4.2%) had this combination with a permanent tracheostomy performed. Nares resection and staphylectomy/palatoplasty were performed in 4/24 cases (16.7%). Staphylectomy/palatoplasty and laryngeal saccule excision were performed in 2/24 cases (8.3%). One isolated case (4.2%) had laryngeal saccule excision alone, and 1 isolated case (4.2%) had a staphylectomy performed alone.

Radiographic dimensions

Orientation (left versus right lateral) was considered in the analysis by including it as a factor in the analysis. Neither the main effect nor any interaction with the other factors was significant for TT:3R measurements (P = 0.63) and TD:TI measurements (P = 0.92).

TD:TI dimensions

Group A (Postoperative radiographs ≤ 1 wk)

When brachycephalic airway cases that had postoperative thoracic radiographs ≤ 1 wk were evaluated, there was no significant difference (P = 0.06) in the mean TD:TI ratios when comparing preoperative [mean ± standard error (SE): 0.13 ± 0.01] to postoperative (mean ± SE: 0.11 ± 0.01) ratios of lateral thoracic radiographs.

Group B (Postoperative radiographs > 1 wk)

When brachycephalic airway cases that had postoperative thoracic radiographs > 1 wk were evaluated, there was no significant difference (P = 0.06) in the mean TD:TI ratios when comparing preoperative (mean ± SE: 0.12 ± 0.01) to postoperative (mean ± SE: 0.14 ± 0.01) ratios of lateral thoracic radiographs.

TT:3R dimensions

Group A (Postoperative radiographs ≤ 1 wk)

When brachycephalic airway cases that had postoperative thoracic radiographs ≤ 1 wk were evaluated, there was no significant difference (P = 0.19) in the mean TT:3R ratios when comparing preoperative (mean ± SE: 2.36 ± 0.14) to postoperative (mean ± SE: 2.11 ± 0.14) ratios of lateral thoracic radiographs.

Group B (Postoperative radiographs > 1 wk)

When brachycephalic airway cases that had postoperative thoracic radiographs > 1 wk were evaluated, there was no significant difference (P = 0.32) in the mean TT:3R ratios when comparing preoperative (mean ± SE: 2.42 ± 0.12) to postoperative (mean ± SE: 2.60 ± 0.12) ratios of lateral thoracic radiographs.

Tracheal hypoplasia

In this study tracheal hypoplasia was diagnosed using both the TD:TI and TT:3R methods (< 0.12 in bulldogs and < 0.16 in non-bulldog brachycephalic breeds for the TD:TI; and < 2.0 for the TT:3R method). Table 1 shows the number of dogs (English bulldog versus other breeds) for group A that were diagnosed with tracheal hypoplasia using the TD:TI measurement method both before surgery (17/24) and after surgery (17/24). This table also shows the number of dogs for group A that were diagnosed with tracheal hypoplasia before surgery (10/24) and after surgery (10/24) using the TT:3R measurement method. Table 2 shows the number of dogs for group B that were diagnosed with tracheal hypoplasia before surgery (25/30) and after surgery (7/30) using the TD:TI measurement method. This table also shows the number of dogs for group B that were diagnosed with tracheal hypoplasia before surgery (7/30) and after surgery (5/30) using the TT:3R measurement method.

Table 1.

Number of dogs (English bulldog versus other breeds) for group A (postoperative radiographs ≤ 1 week after surgery) that were diagnosed with tracheal hypoplasia before and after surgery using the TD:TI measurement method and TT:3R measurement method both before and after surgery.

Group A	TD:TI method for diagnosis of hypoplasia		TT:3R method for diagnosis of hypoplasia
	English bulldog (n = 5) TD:TI < 0.12	Other breeds (n = 7) TD:TI < 0.16	English bulldog (n = 5) TT:3R < 2.0	Other breeds (n = 7) TT:3R < 2.0
Pre right lateral	4	5	3	2
Pre left lateral	3	5	3	2
Post right lateral	3	6	2	3
Post left lateral	2	6	2	3

Table 2.

Number of dogs for group B (postoperative radiographs > 1 week after surgery) that were diagnosed with tracheal hypoplasia before surgery and after surgery using the TD:TI and TT:3R measurement methods.

Group B	TD:TI method for diagnosis of hypoplasia		TT:3R method for diagnosis of hypoplasia
	English bulldog (n = 5) TD:TI < 0.12	Other breeds (n = 10) TD:TI < 0.16	English bulldog (n = 5) TT:3R < 2.0	Other breeds (n = 10) TT:3R < 2.0
Pre right lateral	4	9	2	1
Pre left lateral	3	9	2	2
Post right lateral	2	6	2	1
Post left lateral	2	8	2	0

Discussion

Brachycephalic airway syndrome is comprised of a number of anatomic changes which lead to increased inspiratory resistance (17). The nose is the greatest source of airway resistance within the upper respiratory system, and stenotic nares are the most common primary manifestation of BAS, found in 17% to 77% of brachycephalic breeds (4,18,19). Despite this, the overlong soft palate has also been considered the main cause of BAS clinical signs. More recently, aberrant nasopharyngeal turbinates, which have been found in 21% of brachycephalic dogs (18), have been postulated to play a major role in BAS and have also been shown to be present in clinically healthy English bulldogs (3). However, this needs to be more thoroughly evaluated and was beyond the scope of this study (20–22). A recent study using computational fluid dynamics showed that the rostral third of the nasal passage exhibited a larger airflow resistance than the caudal and middle regions of the nasal passage (23). Hypoplastic trachea is a less common finding (13% of all dogs with BAS; 1,18) and also contributes to increased airway resistance, owing to the decreased tracheal diameter. Hypoplastic trachea has been diagnosed in 53.9% of English bulldogs surgically treated for BAS in previous studies (1); this is similar to an average of 70% of English bulldogs before surgery in this study using the TD:TI method and 50% of English bulldogs using the TT:3R method.

English bulldogs (61%), pugs (21%), and Boston terriers (9%) are the most common breeds with BAS (19). The most common components of BAS were elongated soft palate (94%), stenotic nares (77%), everted laryngeal saccules (66%), and everted tonsils (56%) in a study by Fasanella et al (19). Dogs most commonly had 3 or 4 components of BAS, with the most common combination being stenotic nares, elongated soft palate, everted laryngeal saccules, and everted tonsils (19). Riecks et al (1) concluded that surgical treatment of BAS in dogs appeared to have a favorable long-term outcome, regardless of age, breed, specific diagnosis, or number and combination of diagnoses, including hypoplastic trachea.

In the present study, an elongated soft palate was surgically treated either by a traditional staphylectomy or a folded flap palatoplasty (FFP) and was the most common procedure performed (23/24 cases). This was determined by surgeon preference with the traditional cut and sew staphy lectomy performed to address the elongation of the soft palate (17,24). The FFP has been used to address the soft palate hyperplasia and excessive length, but there is no information demonstrating that one technique is superior to another as far as clinical outcome (25,26).

In this study, everted laryngeal saccules, a secondary component of BAS, were excised in 19/24 cases. Whether or not resection of everted laryngeal saccules (ELS) is needed still remains to be determined. In recent studies in which stenotic nares and elongated palates were corrected and ELS were not addressed, outcomes appeared to be similar to studies in which ELS were excised (25,27). Recent studies have also suggested that sacculectomy may increase morbidity following brachycephalic airway surgery (28).

Tracheal hypoplasia is a congenital abnormality often identified in BAS patients, and the English bulldog has the highest incidence of tracheal hypoplasia among brachycephalic breeds (1,6,14,18,27,29–31). Tracheal hypoplasia is a primary component of BAS (18) along with other abnormalities including stenotic nares, elongated soft palate, redundant pharyngeal tissue, and aberrant nasal conchae. With canine tracheal hypoplasia the tracheal cartilages are small and rigid and the ends are closely apposed or overlapping with shortening of the dorsal elastic membrane and the trachealis muscle (7,10,11,14,30,32). Unlike tracheal collapse, the trachea luminal diameter is reduced with tracheal hypoplasia but does not vary with dynamic pressure changes during respiration (6,30,33).

Although tracheal hypoplasia is considered a significant finding in dogs with clinical BAS, there may be normal variation in tracheal dimensions in healthy dogs of different breeds with BAS (15). There is currently no information on what degree of hypoplasia is tolerated or what degree may cause respiratory symptoms in brachycephalic breeds. Tracheal hypoplasia has been reported to not cause clinical signs and, in fact (1,6,10) may be an incidental finding (14).

Tracheal hypoplasia has historically been thought to contribute to respiratory symptoms in BAS dogs, but studies have not been able to associate tracheal hypoplasia with worse outcomes following surgical treatment of nasal, pharyngeal, and laryngeal abnormalities in dogs. In the absence of concurrent pulmonary disease or cardiovascular disease, tracheal hypoplasia is well tolerated and is not a contraindication for surgery (6). A recent study revealed partial resolution of hypoplastic trachea in 6 English bulldog puppies with bronchopneumonia and determined that tracheal hypoplasia might partially or completely resolve with growth to mature body size (14). Although tracheal hypoplasia is considered a congenital trait, there is evidence that tracheal diameters can change with resolution of bronchopneumonia, during phases of respiration, and with patient growth, demonstrating that tracheal diameter may not be static (14,15). However, based on the present retrospective study, surgical correction of BAS does not change the incidence of tracheal hypoplasia after surgery compared to before surgery.

The 2 methods most widely referred to for evaluation of tracheal dimensions are the tracheal diameter to thoracic inlet (TD:TI) ratio (11) and the thoracic trachea diameter to width of the third rib (TT:3R) ratio described by Suter et al (7) and subsequently modified by Coyne and Fingland (6). The TD:TI ratio is used most commonly to define tracheal hypoplasia and is defined as < 0.16 in non-bulldog brachycephalic dogs and < 0.12 in English bulldogs. For non-brachycephalic dog breeds, normal radiographic tracheal dimensions have been described as TD:TI ratio of > 0.2 (11). Previously reported definitions of tracheal hypoplasia using TT:3R ratios were either < 2.0 or < 3.0 (6–8).

Ingman et al (13) evaluated both the TD:TI and TT:3R methods and concluded that there was poor agreement in classification of English bulldogs as tracheal hypoplastic or non-hypoplastic depending on measuring method, cut-off value and observer (13). In our study, we used both methods for tracheal measurements as both methods are commonly referred to and used in clinical practice. We had a single Board-certified radiologist (FVG) perform all measurements to eliminate variability in interobserver agreement. In this study, there was no significant difference in incidence or diagnosis of tracheal hypoplasia when comparing preoperative versus postoperative thoracic radiographs when using either radiographic measurement at any time point.

Limitations of this study include the retrospective nature with a small number of patients included. These patients had a wide range of ages, which was a variable in the study and may account for changes in tracheal dimensions as tracheal size may change with growth. Information was gathered from medical records, and due to the relatively small study population, it may be challenging to reach conclusions regarding findings. Along with the retrospective nature, the surgical procedures performed for each patient were variable and were dependent on surgeon preference, which may be a limitation for outcome assessment. Three of these patients also had a tracheostomy performed, which has the potential to impact long-term tracheal diameter. Prospective, controlled studies are warranted to further investigate whether tracheal hypoplasia persists or improves after surgical correction of BAS. Tracheal dimensions were obtained from radiographs to diagnose tracheal hypoplasia. Tracheoscopy is likely more sensitive for making a definitive diagnosis and also permits gradation of tracheal hypoplasia. A recent study (34) concluded that CT and radiographic tracheal diameter measurement were comparable in English bulldogs; however, diameters for both imaging techniques were not comparable with tracheoscopy scores, and tracheoscopy identified hypoplastic changes in all study dogs. Montgomery et al (35) compared radiography and CT for determining tracheal diameter and length in canine cadavers for the purpose of tracheal stent size selection and results revealed that tracheal measurements via CT were on average 1 mm larger compared with radiographic measurements, which may be important for the determination of stent size but is likely minimal in regards to diagnosis of tracheal hypoplasia. Another limitation of this study is that radiographic measurements for Group B included a wide time period for > 1 wk postoperative radiographs, and tracheal dimensions may have changed given more time after surgical treatment.

In conclusion, there was no difference in the incidence of tracheal hypoplasia before surgery compared to after surgery, based on tracheal dimensions measured on thoracic radiographs. According to these findings, surgical correction for BAS does not resolve or improve tracheal hypoplasia in brachycephalic breeds. CVJ