Tracheal collapse is a chronic respiratory condition commonly diagnosed in small- and toy-breed dogs.1–4 Dynamic mechanical collapse, the traditionally considered form of tracheal collapse, can be caused by a combination of progressive malacia of the tracheal cartilages and laxity of the trachealis muscle, resulting primarily in dorsoventral collapse. However, other varieties of dynamic or static airway obstruction have been identifed.5 A W-shaped malformation of the tracheal cartilages (referred to herein as MTC) can cause a more static form of narrowing of the tracheal lumen dorsoventrally that is distinct from, or in addition to, dynamic narrowing caused by invagination of the trachealis muscle during respiration.5 Clinical signs most consistent with tracheal collapse in dogs include a goose-honking cough, persistent raspy breathing, and episodes of respiratory distress and cyanosis often following excitement that can culminate in life-threatening dyspnea.
Recommended initial management of tracheal collapse in dogs is conservative in nature and includes weight loss, avoidance of neck leads, management of contributing comorbidities, and use of various medications to reduce the severity and frequency of respiratory distress caused by airway collapse.1–3 Given the reported complications associated with tracheal surgery and general anesthesia in these compromised patients, surgery is typically reserved for dogs with refractory, unresponsive, or severe tracheal collapse or dogs that are medically unstable.1–5 Surgical options include the use of extraluminal ring prosthetics or endoluminal stents to reestablish the airway.6–14 Although use of either device can provide an improvement in respiratory function, tracheal stents are often preferred because these devices can be placed quickly and noninvasively along the entire length of the trachea. In addition, endoluminal stent placement can avoid the potentially life-threatening perioperative complications associated with extraluminal prosthetics such as laryngeal paralysis, airway obstruction, or death.6–14 However, the potential for complications also exists with stent placement, including tissue ingrowth, stent migration, stent shortening, stent fracture, inflammatory and bacterial tracheitis, and aspiration pneumonia presumably caused by stent-related alterations in respiratory function.12–21 Although many of these complications can be successfully managed medically with long-term favorable outcomes, serious consequences can result.12–21
Previously reported studies11,14,19,21 of tracheal stent placement in dogs have involved small numbers of patients evaluated over fairly brief periods, thereby providing limited insight into the relative frequency of complications encountered over the long term in larger patient groups. Furthermore, to the authors’ knowledge, no investigation has been made into whether and how complications or prognosis may differ between dogs with TTC versus MTC. Such information may be useful for optimizing case selection and patient outcome, both of which are desirable in the palliative management of dogs with tracheal collapse.
The purpose of the study reported here was to investigate and characterize complications and short-, intermediate-, and long-term outcomes for dogs with tracheal collapse treated by tracheal stent placement by clinicians at a single interventional radiology service. We hypothesized that the associated clinical signs would be improved immediately following stent placement, complication types and frequencies would differ between dogs with TTC and dogs with MTC, and the presence of mainstem bronchial collapse identified at the time of stent placement would have no significant adverse association with patient survival times or rates.
Materials and Methods
Case selection
All dogs with tracheal collapse evaluated by clinicians of the Animal Medical Center interventional radiology service from September 1, 2009, to August 3, 2015, were considered for inclusion in the study. Dogs that underwent 1 or more tracheal stent procedures were included. Dogs were excluded if they had a previous tracheal stent placed elsewhere or history of previous tracheal surgery (eg, a tracheal ring prostheses procedure). Dogs that survived for 90 days after stent placement were required to have received ≥ 90 days of follow-up to permit sufficient time for complications to develop; dogs that died prior to that point but that met the other criteria were included.
Whether tracheal stent placement would be performed was determined on a case-by-case basis. Delaying of this procedure was strongly considered when aggressive multimodality medical management had not yet been attempted or contributing comorbidities had not been adequately managed; however, in uncommon instances, the severity of respiratory compromise superseded these considerations.
Medical records review
Data collected from medical records of dogs meeting the inclusion criteria were grouped into 2 clinically relevant periods: the perioperative period and the postdischarge follow-up period. Throughout the study, dog owners were asked to score the severity of clinical signs by responding to an unvalidated, orally administered questionnaire specifically designed for the study to assess the respiratory status and quality of life in dogs with tracheal collapse. Clinical severity was scored from 0 to 10 for each of 3 clinical signs: goose-honking or raspy breathing, coughing, and dyspnea. Owners were informed that a score of 0 indicated that a clinical sign was never witnessed. Conversely, a score of 10 indicated that a clinical sign was persistent and intractable. For statistical analysis, any scores provided to 1 or more decimal places were rounded to whole numbers. Owners were first asked to score their dogs’ clinical signs during the week prior to the initial appointment with the interventional radiology service. These scores were then obtained at every subsequent appointment with the service and were used to make recommendations regarding medical management and other interventions.
Because of variation in the frequency of postoperative patient visits, reported scores were grouped according to a recommended (rather than actual) appointment schedule relayed to clients orally and via discharge instructions after stent placement. Per this schedule, patients were recommended to return approximately 1 month following stent placement and then every 3 months subsequently. For patients with > 1 visit during each schedule period, mean scores for each clinical sign were calculated. Some patients were reevaluated within each schedule period, and others within a few, or a single, schedule period. All dogs were ultimately included in a final follow-up evaluation category, regardless of when or how often they were reevaluated.
Perioperative period—Perioperative data included any historical patient information as well as information pertaining to the admission event during which the dog received the initial tracheal stent (ie, baseline data), and included the period between hospital admission and hospital discharge or death, whichever occurred first. Specific types of data included patient signalment and pertinent physical examination findings, admitting service, previous medical management for TC, baseline clinical severity scores, pre- and intraoperative imaging findings (thoracic radiography, fluoroscopy, tracheoscopy, and echocardiography, as applicable), tracheal collapse type and location, presence of mainstem bronchial collapse or compression, stent type and dimensions, procedure durations, types of ancillary procedures performed, types of intraoperative and early postoperative complications, whether the patient survived to hospital discharge, cause of death if applicable, and duration of postoperative hospital stay.
Tracheal collapse type was classified as primarily TTC (dorsal membrane laxity, tracheal ring weakness presumptively secondary to chondromalacia, or both) or MTC (W-shaped cartilage rings) when possible (Figure 1). Potential intraoperative complications included anesthetic or stent-associated complications such as stent mis-sizing or misplacement. Potential perioperative complications included pneumonia or tracheal infection (confirmed via endotracheal lavage or radiographic findings), respiratory distress or dyspnea, the need for additional airway surgery or procedures, and death. Perioperative complications were considered major (life-threatening) if a second anesthetic procedure was required or euthanasia was performed and minor (non–life-threatening) if only additional monitoring, a change in medical management, or an extended hospital stay was required.
Radiographic, endoscopic, and postmortem photographic images of 2 dogs, each with a with different type of tracheal collapse (TTC [dorsal membrane laxity, tracheal ring weakness presumptively secondary to chondromalacia, or both; A, B, and C] or MTC [W-shaped cartilage rings; D, E, and F]). A—Lateral radiographic view showing dynamic ventral deviation (arrows) of the weakened, flaccid trachealis muscle. B—Endoscopic image of flaccid trachealis muscle compromising tracheal lumen patency. C—Postmortem lateral photograph of the larynx and cervical portion of the trachea, showing flattened tracheal cartilage rings with a minimal dorsoventral tracheal diameter. D—Lateral radiographic view showing static dorsal deviation and narrowing (arrows) of ventral cartilage rings at the thoracic inlet compromising the tracheal lumen. E—Endoscopic image of static W-shaped cartilage rings obstructing the tracheal lumen at the level of the thoracic inlet. F—Postmortem lateral photograph of the trachea, showing ventral tracheal deviation into the tracheal lumen due to static, malformed, W-shaped cartilage rings.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Radiographic, endoscopic, and postmortem photographic images of 2 dogs, each with a with different type of tracheal collapse (TTC [dorsal membrane laxity, tracheal ring weakness presumptively secondary to chondromalacia, or both; A, B, and C] or MTC [W-shaped cartilage rings; D, E, and F]). A—Lateral radiographic view showing dynamic ventral deviation (arrows) of the weakened, flaccid trachealis muscle. B—Endoscopic image of flaccid trachealis muscle compromising tracheal lumen patency. C—Postmortem lateral photograph of the larynx and cervical portion of the trachea, showing flattened tracheal cartilage rings with a minimal dorsoventral tracheal diameter. D—Lateral radiographic view showing static dorsal deviation and narrowing (arrows) of ventral cartilage rings at the thoracic inlet compromising the tracheal lumen. E—Endoscopic image of static W-shaped cartilage rings obstructing the tracheal lumen at the level of the thoracic inlet. F—Postmortem lateral photograph of the trachea, showing ventral tracheal deviation into the tracheal lumen due to static, malformed, W-shaped cartilage rings.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Radiographic, endoscopic, and postmortem photographic images of 2 dogs, each with a with different type of tracheal collapse (TTC [dorsal membrane laxity, tracheal ring weakness presumptively secondary to chondromalacia, or both; A, B, and C] or MTC [W-shaped cartilage rings; D, E, and F]). A—Lateral radiographic view showing dynamic ventral deviation (arrows) of the weakened, flaccid trachealis muscle. B—Endoscopic image of flaccid trachealis muscle compromising tracheal lumen patency. C—Postmortem lateral photograph of the larynx and cervical portion of the trachea, showing flattened tracheal cartilage rings with a minimal dorsoventral tracheal diameter. D—Lateral radiographic view showing static dorsal deviation and narrowing (arrows) of ventral cartilage rings at the thoracic inlet compromising the tracheal lumen. E—Endoscopic image of static W-shaped cartilage rings obstructing the tracheal lumen at the level of the thoracic inlet. F—Postmortem lateral photograph of the trachea, showing ventral tracheal deviation into the tracheal lumen due to static, malformed, W-shaped cartilage rings.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Stent placement procedure
When possible, dogs were initially treated with aggressive medical management (a combination of antitussive drugs, anti-inflammatory drugs, sedatives or tranquilizers, and bronchodilators given regularly and often in higher than typical dosages, as prescribed and not on an as-needed basis) for a period of weeks to years following diagnosis of tracheal collapse. The duration of this aggressive management was typically recommended to be at least 2 to 4 weeks prior to stent placement to allow identification of responders not requiring surgical intervention.
Some variation in the stent placement procedure occurred over the approximate 5-year study period; however, the basic procedure was performed with slight variation of that previously described5 (Figure 2). Generally, premedication was provided before anesthetic induction to facilitate IV catheter placement. An anti-inflammatory dose of corticosteroid medication was administered perioperatively, provided the dog had not already received the medication that day. The dog was positioned in sternal recumbency, anesthesia was lightly induced by IV administration, and oral and laryngeal examinations were performed. Additional anesthetics were administered as necessary to facilitate subsequent flexible tracheoscopy. Immediately after tracheoscopy, the dog was intubated with a sterile endotracheal tube. It was then positioned in lateral recumbency, the neck was ventroflexed so the trachea was straightened, and the endotracheal tube was positioned such that the cuff was immediately caudal to the cricoid cartilage.
Lateral thoracic fluoroscopic images of a dog with tracheal collapse before (A) and immediately after (B) endoluminal placement of a tracheal stent. An esophageal marker catheter has been placed to calculate radiographic magnification. A—Positive-pressure ventilation at 20 cm H2O was used to achieve tracheal distension, allowing measurement of maximal tracheal diameters in the cervical, thoracic inlet, and intrathoracic regions by use of an esophageal marker catheter. B —Appropriate stent placement and ultimate tracheal stent dimensions are confirmed.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Lateral thoracic fluoroscopic images of a dog with tracheal collapse before (A) and immediately after (B) endoluminal placement of a tracheal stent. An esophageal marker catheter has been placed to calculate radiographic magnification. A—Positive-pressure ventilation at 20 cm H2O was used to achieve tracheal distension, allowing measurement of maximal tracheal diameters in the cervical, thoracic inlet, and intrathoracic regions by use of an esophageal marker catheter. B —Appropriate stent placement and ultimate tracheal stent dimensions are confirmed.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Lateral thoracic fluoroscopic images of a dog with tracheal collapse before (A) and immediately after (B) endoluminal placement of a tracheal stent. An esophageal marker catheter has been placed to calculate radiographic magnification. A—Positive-pressure ventilation at 20 cm H2O was used to achieve tracheal distension, allowing measurement of maximal tracheal diameters in the cervical, thoracic inlet, and intrathoracic regions by use of an esophageal marker catheter. B —Appropriate stent placement and ultimate tracheal stent dimensions are confirmed.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
A hydrophilic guidewirea and marker catheterb combination was advanced down the esophagus and positioned across the length of the trachea. Positive-pressure ventilation of 20 cm H2O was performed through the endotracheal tube, and a radiographic image was obtained to establish maximal tracheal dimensions. Negative-pressure ventilation of −10 to −15 cm H2O was then performed to identify the extent of the tracheal and mainstem bronchial collapse present. The esophageal marker catheter was used to calibrate the radiographic magnification.
A tracheal SEMSc,d with a diameter approximately 2 to 3 mm greater than the largest tracheal diameter was chosen. A tracheal stent–shortening chart provided by the stent manufacturer was used to calculate the appropriate stent length such that the stent would extend from approximately 10 mm caudal to the cricoid cartilage to approximately 10 mm cranial to the carina. The stent was then placed under fluoroscopic guidance through the endotracheal tube, and the tube cuff was occasionally used to further expand the stent if necessary, particularly for malformations that in the authors’ experience often limited stent expansion and tracheal wall contact. If necessary, additional stents were placed to increase the length of trachea covered (dogs with any type of tracheal collapse) or achieve greater expansile forces (dogs with MTC).
After stent placement, the dog was repositioned in sternal recumbency and extubated, and flexible tracheoscopy was repeated to confirm appropriate positioning and size of the stent as well as contact of the stent with the tracheal wall. If the stent placement was considered inappropriate or insufficient, the surgeon removed the stent under fluoroscopic guidance using alligator grasping forceps to grab the rostroventral aspect of the stent and replaced it with a more appropriately sized stent. A sterile endotracheal tube was then used for reintubation, and endotracheal lavage was performed to collect fluid samples for cytologic examination as well as bacterial culture and antimicrobial susceptibility testing. Additional surgical procedures such as partial staphylectomy or epiglottopexy were then performed if deemed necessary.
After recovering from anesthesia, the dog was moved to the intensive care unit overnight, where it received antitussive and corticosteroid drugs and antimicrobials. Thoracic radiography (standard 3 views, including the neck) was typically performed the following day, prior to hospital discharge. Antimicrobials were routinely administered for 10 to 14 days unless pneumonia was suspected or confirmed via radiography or results of endotracheal fluid testing, in which case appropriate antimicrobials were administered for 4 to 6 weeks. The dosage of corticosteroid drugs administered was typically reduced slowly over the first 1 to 3 months when possible. Recheck appointments for examination and thoracic-cervical radiography were scheduled at 1 month and then every 3 months after stent placement or sooner if problems occurred.
Follow-up after hospital discharge
Follow-up information was collected on each dog via recheck appointment or telephone contact with the referring veterinarian or owner. Postdischarge data included information collected between hospital discharge (after initial tracheal stent placement) and the final follow-up examination or death and were categorized as short term (< 45 days after stent placement), intermediate term (45 to 180 days after stent placement), or long term (> 180 days after stent placement). Collected information included severity of clinical signs (scores), medications administered, minor and major complications, survival time, and cause of death when applicable. Potential postoperative complications included stent fracture (loss of conformation or integrity along the stent body), inflammatory or granulation tissue ingrowth within the tracheal lumen (confirmed by radiographic, ultrasonographic, endoscopic, or postmortem findings), pneumonia or tracheal infection, progressive tracheal collapse, and stent migration or shortening. Stent fraying, defined as small wire fractures at either end of the stent, was not considered a complication because it is the authors’ opinions from years of monitoring these patients that this occurrence does not clearly result in further stent-related complications (Figure 3). Identified complications were categorized as minor (non–life-threatening, requiring monitoring or medical management) or major (requiring an additional procedure or resulting in death or euthanasia). When possible, the cause of death was categorized as confirmed via postmortem examination to be due the stent or to tracheal collapse, suspected to be due to the stent or tracheal collapse, unlikely to be due to the stent or tracheal collapse, or confirmed via postmortem examination or medical records review to not be due to the stent or tracheal collapse.
Lateral thoracic radiographic images of 3 dogs that underwent endoluminal tracheal stent placement for treatment of tracheal collapse. A—Intact tracheal stent with no evidence of fraying (ie, rounded wire edges of the stent [arrows]) or fracture. B—Tracheal stent with evidence of stent fraying at the cranial edge (arrows). C—Tracheal stent with evidence of early fracture along the dorsal, intrathoracic portion of the stent (arrows). This stent also has fraying along its caudal margin.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Lateral thoracic radiographic images of 3 dogs that underwent endoluminal tracheal stent placement for treatment of tracheal collapse. A—Intact tracheal stent with no evidence of fraying (ie, rounded wire edges of the stent [arrows]) or fracture. B—Tracheal stent with evidence of stent fraying at the cranial edge (arrows). C—Tracheal stent with evidence of early fracture along the dorsal, intrathoracic portion of the stent (arrows). This stent also has fraying along its caudal margin.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Lateral thoracic radiographic images of 3 dogs that underwent endoluminal tracheal stent placement for treatment of tracheal collapse. A—Intact tracheal stent with no evidence of fraying (ie, rounded wire edges of the stent [arrows]) or fracture. B—Tracheal stent with evidence of stent fraying at the cranial edge (arrows). C—Tracheal stent with evidence of early fracture along the dorsal, intrathoracic portion of the stent (arrows). This stent also has fraying along its caudal margin.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Statistical analysis
Because the study was intended as exploratory in nature, no a priori sample size calculation was performed. Descriptive statistics were computed for all variables. Between-group comparisons involving continuous variables were performed by means of ANOVA or the Wilcoxon rank sum test, as appropriate for the data distribution. Between-group comparisons involving categorical variables were performed with the χ2 test or with the Fisher exact text for cells containing < 5 observations. For statistical purposes, dogs admitted through services other than interventional radiology and emergency were grouped with those admitted through the interventional radiology service in relevant comparisons.
Other analyses were performed to investigate associations between several variables (dog age, sex, and breed [Yorkshire Terrier vs other dogs]; type of tracheal collapse [MTC vs TTC]; perioperative pneumonia or tracheal infection [yes vs no]; mainstem bronchial collapse or compression [yes vs no]; history of cardiac abnormalities [yes vs no]; type of major complications; and number of stent placement procedures) and survival time or rate. Univariate time-to-event analyses were performed with the Kaplan-Meier method to account for right censoring of dogs lost to follow-up; a log-rank P value was used to assess between-group differences. Cox semiparametric proportional hazards models were used to generate hazard ratios for univariate and multivariate models. Tests for proportionality were carried out by visual inspection of Schoenfeld residuals, the negative logarithmic-estimated spectral density function, and formal hypothesis testing and deemed proportional.
All analyses were performed with standard statistical software.e Values of P < 0.05 were considered significant. No adjustment was made for testing of multiple hypotheses.
Results
Dogs
Review of the medical records revealed 157 dogs with tracheal collapse that were examined by clinicians of the Animal Medical Center interventional radiology service during the study period. Of those 157 dogs, 85 (54%) received endoluminal tracheal stents. However, only 76 had stents initially placed by service clinicians. One of these patients had tracheal ring prostheses placed previously elsewhere, leaving 75 of 157 (48%) dogs for inclusion in the study.
Seven dog breeds were represented, the most common of which were Yorkshire Terrier (67% [50/75]) and Pomeranian (17% [13/75]). There were also 2 (3%) mixed-breed dogs. Males (55% [41/75]) and females (45% [34/75]) ranged in age from 2.3 to 15.2 years (mean ± SD, 8.2 ± 3.4 years), weighed between 1.1 and 9.2 kg (2.4 to 20.2 lb; mean ± SD, 3.5 ± 1.3 kg [7.7 ± 2.9 lb]), and had body condition scores between 3/9 and 8/9 (median, 5/9). Median age of males (8.3 years) and females (8.1 years) did not differ significantly (P = 0.80). Various comorbidities were identified in historical medical records, such as but not limited to cardiac abnormalities (28% [21/75]), which included cardiomegaly (21% [16/75]) or heart murmur (15% [11/75]), and pneumonia (12% [9/75]).
Fifty- one percent (38/ 75) of dogs had MTC and 49% (37/75) had TTC. The proportion of males and females with MTC was similar (54% [22/41] and 47% [16/34], respectively; P = 0.65). Characteristics of included dogs were summarized by type of tracheal collapse (Table 1) and breed (Yorkshire Terriers vs other dogs; Supplementary Table S1, available at avmajournals.avma.org/doi/suppl/10.2460/javma.254.3.380).
Characteristics of 75 dogs with tracheal collapse treated by endoluminal tracheal stent placement, grouped by type of tracheal collapse.
| Characteristic | TTC (n = 37) | MTC (n = 38) | P value |
|---|---|---|---|
| Age (y) | 9.4 ± 3.5 | 7.1 ± 2.8 | 0.002 |
| Yorkshire Terrier | 18 (49) | 32 (84) | 0.003 |
| Male | 20 (54) | 22 (58) | 0.92 |
| Body weight (kg) | 3.69 ± 1.07 | 3.25 ± 1.48 | 0.11 |
| Cardiac abnormality* | 13 (35) | 8 (21) | 0.27 |
| Emergency service admission† | 21 (57) | 18 (47) | 0.56 |
| No. of stent placement procedures | 1.41 ± 0.60 | 1.63 ± 0.71 | 0.15 |
| Major postoperative complication‡ | 12 (35) | 21 (58) | 0.08 |
| Stent fracture | 7 (21) | 6 (17) | 0.96 |
| Tissue ingrowth | 4 (12) | 8 (22) | 0.37 |
| Progressive tracheal collapse | 1 (3) | 6 (17) | 0.11 |
| Stent dislodgement | 0 (0) | 1 (3) | 1.00 |
Data for age, body weight, and number of stent placement procedures are reported as mean ± SD; all other data represent number (%) of dogs with the indicated characteristic.
Includes dogs with a history of cardiac abnormality including cardiomegaly and heart murmur (vs dogs with no history of cardiac abnormality).
Dogs were admitted to the hospital through the emergency service, interventional radiology service, or another service.
Includes major complications occurring any time after stent placement and requiring deployment of another stent; 34 TTC patients and 36 MTC patients survived to hospital discharge and are included in these calculations.
Perioperative period
Fifty-two percent (39/75) of dogs were admitted to the hospital through the emergency service, 45% (34/75) through the interventional radiology service, and 3% (2/75) through other services. For statistical purposes, the dogs admitted through other services were grouped with the dogs admitted through the interventional radiology service.
Thirty-five percent (26/75) of dogs had been aggressively medically managed specifically by clinicians of the interventional radiology service before the stent placement procedure was performed. The duration of clinical signs was < 1 year for 29% (20/69) of dogs with available information, between 1 and 3 years for 44% (30/69), and > 3 years for 28% (19/69). Median baseline clinical severity scores for goose-honking or raspy breathing, coughing, and dyspnea were summarized (Table 2). Among dogs with available information, 85% (51/60) were regularly receiving medical treatment for tracheal collapse immediately before stent placement and 15% (9/60) were not. The most commonly used medications for those that received medical management were antitussive drugs (92% [47/51]) and corticosteroid drugs (73% [37/51]). The duration of medical management for these 51 dogs ranged from 3 days to 2,391 days (median, 207 days). No significant (P = 0.08) difference in duration of prior medical management was identified between dogs with MTC (mean ± SD, 568 ± 599 days) and those with TTC (mean ± SD, 287 ± 375 days).
Median clinical severity scores (No. of dogs with data) before (baseline) and at various points after tracheal stent placement for the dogs of Table 1.
| Clinical sign | Baseline | Short term | Intermediate term | Long term | Final follow-up | P value* |
|---|---|---|---|---|---|---|
| Coughing | 2 (57) | 2.3 (42) | 2.5 (35) | 2.9 (41) | 2.3 (60) | 0.15 |
| Goose-honking or raspy breathing | 7 (54) | 0.5 (37) | 0 (33) | 1.7 (40) | 1 (57) | < 0.001 |
| Dyspnea | 6.5 (57) | 0 (41) | 0 (35) | 0.5 (41) | 0 (59) | < 0.001 |
P values represent comparisons of the mean final follow-up score with the mean baseline score.
Scores from 0 to 10 were assigned by dog owners; 0 indicates the absence of a clinical sign, and 10 indicates persistence and constancy of that sign. Short term was defined as < 45 days after stent placement, intermediate term as 45 to 180 days after stent placement, and long term as > 180 days after stent placement.
Preoperative thoracic radiographs were available for review for 87% (65/75) of dogs. Various comorbidities involving the respiratory, cardiac, hepatic, skeletal, and gastrointestinal systems were identified. Some of the more commonly recognized comorbidities included cardiac enlargement (33% [25/75]), hepatomegaly (33% [25/75]), and rib fractures (8% [6/75]). Echocardiography was performed some time prior to stent placement for 24% (18/75) of dogs.
Stent placement procedure
Fifty-nine percent (44/75) of dogs had tracheoscopy performed, and all had fluoroscopy performed. The location of tracheal collapse was identified as solely cervical, thoracic inlet, and intrathoracic in 4% (3/75), 12% (9/75), and 3% (2/75) of dogs, respectively. Collapse was considered to exist in 2 locations only in 17% (13/75) of dogs and was identified as diffuse along the entire trachea in 64% (48/75) of dogs. Mainstem bronchial collapse, compression, or narrowing was identified in 70% (43/61) of dogs in which bronchial appearance was described.
Tracheal stents (119 in total) placed in the 75 dogs ranged in diameter from 8 to 18 mm and in length from 32 to 90 mm. Mean ± SD values for stent diameter and length in dogs with MTC were 12.8 ± 1.6 mm and 58.6 ± 10.9 mm, respectively, and for dogs with TTC were 13.2 ± 1.6 mm and 63.1 ± 10.1 mm, respectively (P = 0.34 for diameter comparison and P = 0.08 for length comparison). A nitinol SEMSc was used 90% (107/119) of the time, and a cobalt-chromium-nickel-molybdenum SEMSd was used 10% (12/119) of the time. The most commonly used stents were those with an external diameter and length of 14 mm and 58 mm (34% [41/119]), 14 mm and 72 mm (12% [14/119]), and 12 mm and 52 mm (12% [14/119]), respectively. No significant (P = 0.62) difference in mean stent length was identified between the first 50% of dogs treated (mean ± SD, 61.5 ± 11.9 mm) and the subsequently treated dogs (mean ± SD, 60.3 ± 9.2 mm). However, mean stent diameter increased significantly (P = 0.01) between the first half of dogs treated (mean ± SD, 12.5 ± 1.8 mm) and the subsequently treated dogs (mean ± SD, 13.5 ± 1.3 mm). Observations were similar when limited to only dogs with MTC. No significant difference in in mean stent length (P = 0.39) or mean stent diameter (P = 0.41) by treatment period was identified for dogs with TTC. Intraoperative stent replacement due to initial inappropriate stent sizing was performed for 5% (4/75) of dogs.
Ancillary surgical procedures were performed for 24% (18/75) of dogs. These procedures included partial staphylectomies for elongated soft palates (24% [18/75]), epiglottopexies (8% [6/75]), and bilateral laryngeal sacculectomy (1% [1/75]). Median procedure duration for tracheal stent placement with tracheoscopy and endotracheal lavage but no ancillary procedures was 30 minutes (range, 15 to 85 minutes; n = 51). Median procedure duration for these listed procedures plus ancillary surgical procedures was 35 minutes (range, 15 to 125 minutes; n = 66).
Endotracheal lavage with cytologic evaluation and bacterial culture was performed for 56% (42/75) of dogs, and positive bacterial culture results were obtained for 79% (33/42). Dogs with TTC (n = 37) had 26 bacterial cultures performed at stent placement, 73% (19/26) of which yielded positive results. Dogs with MTC (n = 38) had 16 bacterial cultures performed at stent placement, 88% (14/16) of which yielded positive results. The difference between groups in proportions of dogs with positive culture results at this point was not significant (P = 0.18).
Minor perioperative complications occurred in 55% (41/75) of dogs and included pneumonia or tracheal infection (52% [39/75]), respiratory distress or dyspnea (7% [5/75]), and regurgitation (3% [2/75]). Major perioperative complications occurred in 9% (7/75) of dogs and included death or euthanasia (7% [5/75]) and respiratory distress or dyspnea (3% [2/75]). The 2 dogs with respiratory distress or dyspnea had required additional procedures owing to persistent tracheal collapse caudally; one dog needed an additional stent, laryngeal saccule removal, and tracheostomy, and the other needed partial staphylectomy. Deaths within the perioperative period were attributed to noncardiogenic pulmonary edema (n = 2) or progressive dyspnea due to pneumonia (2), or no cause was determined (1). Ninety-three percent (70/75) of dogs were discharged from the hospital between 0 and 9 days after stent placement (median, 1 day).
Follow-up after hospital discharge
Short-term period (< 45 days after stent placement)—Twenty percent (14/70) of surviving dogs did not return for reevaluation within the short-term follow-up period, leaving 56 (80%) dogs eligible for inclusion in the short-term analysis. Median clinical severity scores and changes in those scores during this period were summarized (Table 2; Supplementary Table S2, available at avmajournals.avma.org/doi/suppl/10.2460/javma.254.3.380). Compared with baseline scores, short-term scores indicated significant improvement in goose-honking or raspy breathing (P < 0.001) and dyspnea (P < 0.001), but not in coughing (P = 0.21). Medical management information was available for review for 52 dogs, all of which were still being medicated in the short-term period. These medications most commonly included antitussive drugs (98% [51/52]), corticosteroid drugs (87% [45/52]), sedatives or antianxiety medications (56% [29/52]), and antimicrobials (54% [28/52]).
Minor complications, identified in 21% (12/56) of dogs evaluated in the short-term period, were limited to medically managed pneumonia or tracheal infections and were identified via radiography alone (n = 9) or both radiography and bacterial culture (3). Major complications were reported for 9% (5/56) of dogs. Two dogs had major stent fractures, 1 dog had persistent tracheal collapse cranial and caudal to the previous stent, 1 dog had persistent tracheal collapse caudal to the previous stent, and 1 dog had a stent dislodged during an anesthetic episode at a primary veterinary clinic.
Additionally, 2 of 56 (4%) dogs died of respiratory-related signs, beginning with dyspnea and progressing to cardiopulmonary arrest. Necropsy of 1 dog confirmed the radiographic diagnosis of severe pneumonia. Necropsy of the other dog was declined; however, this dog had a radiographic diagnosis of pneumonia.
Intermediate-term period (45 to 180 days after stent placement)—Thirty-four percent (23/68) of dogs surviving at least 45 days did not return for reevaluation within the intermediate-term follow-up period, leaving 45 dogs eligible for inclusion in the intermediate-term analysis. Median clinical severity scores during this period and changes in those scores were summarized (Table 2; Supplementary Table S2). Medical management information was available for review for all 45 dogs, all of which were still being medicated in the intermediate-term period. These medications most commonly included antitussive drugs (93% [43/45]), corticosteroid drugs (87% [39/45]), sedatives or antianxiety medications (47% [21/45]), and antimicrobials (44% [20/45]).
Minor complications, identified in 18% (8/45) of dogs evaluated during the intermediate-term period, included medically managed tissue ingrowth (11% [5/45]) and medically managed pneumonia or tracheal infection confirmed via radiography alone (4% [2/45]) or both radiography and bacterial culture (2% [1/45]). Major complications were reported for 9% (4/45) of dogs. Two dogs had stent fractures, 1 dog had progressive tracheal collapse cranial to the previous stent, and 1 dog had tissue ingrowth resulting in airway obstruction. Additionally, 3 of 45 (7%) dogs died of respiratory-related signs during this period and 2 (4%) others died of unknown causes.
Long-term period (> 180 days after stent placement)—Twenty-one percent (13/63) of dogs surviving > 180 days did not return for reevaluation during the long-term follow-up period, leaving 50 dogs eligible for inclusion in the long-term analysis. Median clinical severity scores during this period and changes in those scores were summarized (Table 2; Supplementary Table S2). Medical management information was available for review for 90% (45/50) of dogs, all of which were still receiving medications. These medications most commonly included antitussive drugs (91% [41/45]), corticosteroid drugs (87% [39/45]), antimicrobials (49% [22/45]), and sedatives or antianxiety medications (40% [18/45]).
Minor complications, identified in 66% (33/50) of dogs evaluated during the long-term period, included medically managed pneumonia or tracheal infection (58% [29/50]) and medically managed tissue ingrowth (22% [11/50]). Pneumonia or tracheal infection was confirmed via radiography alone (24% [7/29]), bacterial culture alone (21% [6/29]), or both radiography and bacterial culture (55% [16/29]). Major complications were reported for 48% (24/50) of dogs. Eleven dogs had obstructive tissue ingrowth, 9 dogs had major stent fracture, and 4 had progressive tracheal collapse. Thirteen of 63 (21%) dogs died of respiratory-related signs during this period, 4 (6%) died of nonrespiratory causes, and 13 (21%) died of unknown causes.
Final follow-up examination—Follow-up times after stent placement for all 75 dogs included in the study ranged from 0 (perioperative death) to 2,119 days (median, 728 days). Median clinical severity scores at the final follow-up examination and changes in those scores were summarized (Table 2; Supplementary Table S2). Compared with baseline clinical severity scores, final follow-up examination scores indicated a significant improvement in goose-honking and raspy breathing (P < 0.001) and dyspnea (P < 0.001), but not in coughing (P = 0.15). Eighty-nine percent (42/47) of dogs had an improvement in goose-honking and raspy breathing, 84% (43/51) had an improvement in dyspnea, and only 43% (22/51) had an improvement in coughing. Medical management information was available for review for 85% (64/75) of dogs, which were primarily receiving antitussive drugs (92% [59/64]), corticosteroid drugs (80% [51/64]), antimicrobials (41% [26/64]), and sedatives or antianxiety medications (38% [24/64]).
A single stent placement procedure was performed for 53% (37/70) of dogs that survived to hospital discharge, and 2, 3, and 4 stent placement procedures were performed for 40% (28/70), 6% (4/70), and 1% (1/70) dogs, respectively. These additional procedures were required because of major complications, including stent fracture (19% [13/70]), obstructive tissue ingrowth (17% [12/70]), progressive tracheal collapse (10% [7/70]), and stent dislodgement during early intubation (1% [1/70]). Fifty-six percent (20/36) of dogs with MTC and 38% (13/34) of dogs with TTC required an additional stent placement procedure (P = 0.13). No significant difference in proportions of dogs with each of the aforementioned major complications (thereby requiring an additional stent placement procedure) was identified between dogs with TTC and those with MTC (Table 1). Excluding dogs with progressive tracheal collapse (shorter stents may have initially been placed in these patients), 47% (14/30) of dogs with MTC and 36% (12/33) of dogs with TTC required an additional stent placement procedure (P = 0.57).
The most common minor complications included pneumonia or tracheal infection (57% [40/70]) and tissue ingrowth managed medically (17% [12/70]). Pneumonia or tracheal infection was confirmed via radiography alone (38% [15/40]), bacterial culture alone (15% [6/40]), and both radiography and bacterial culture (48% [19/40]). This complication was identified in 58% (21/36) of dogs with MTC and 56% (19/34) of dogs with TTC. Bacterial cultures were performed for 32 dogs after hospital discharge, of which 25 (78%) had positive results (64% [7/11] of dogs with TTC and 86% [18/21] of dogs with MTC; P = 0.20). Some cultures yielded multiple bacterial species. For dogs that required additional stent placement procedures and had endotracheal lavage samples collected for bacterial culture at that time (24/30 [80%]), positive culture results were not significantly more likely for dogs with (vs without) tissue ingrowth (P = 0.37), progressive tracheal collapse (vs no such collapse; P = 0.30), or stent dislodgement (vs no dislodgment; P = 1.00). However, positive culture results were significantly (P = 0.03) less likely for dogs with (vs without) prior stent fracture.
Only 1 of 13 dogs that required an additional stent placement procedure because of prior stent fracture developed another stent fracture. However, 3 of 12 dogs that required an additional stent placement procedure because of tissue ingrowth following the original procedure required yet another stent placement procedure. Although a greater proportion of dogs with MTC had tissue ingrowth (whether managed medically or by another stent placement procedure; 39% [14/36]) than did dogs with TTC (29% [10/34]), this difference in ingrowth rates was not significant (P = 0.56).
Overall survival rates and causes of death
By the end of the study, 30 of 75 (40%) dogs remained alive and 45 (60%) dogs had died. Overall survival rates were 42% (16/38) for dogs with MTC and 38% (14/37) for dogs with TTC. Mean survival time and MST for the 45 dogs that died were 1,042 and 1,005 days, respectively. Cause of death was determined for 67% (30/45) of dogs and was respiratory related for 77% (23/30) and non–respiratory related for 23% (7/30). For dogs with a respiratory-related death, the cause could not always be differentiated between stent-associated or other respiratory causes.
The estimated MST for all 75 dogs was 1,005 days (Figure 4). Results of Kaplan-Meier analyses comparing survival times between various groups were summarized (Table 3). Median survival times were significantly longer for dogs < 7.7 years of age at the time of stent placement (vs older dogs) and male (vs female) dogs. Dogs that required another stent placement procedure because of stent fracture had a significantly (P = 0.02) shorter survival time than those requiring another procedure because of obstructive tissue ingrowth or progressive tracheal collapse. However, MST for dogs requiring at least 2 stent placement procedures did not differ significantly (P = 0.33) from that of dogs requiring only 1 stent placement procedure. No significant associations with survival time were identified for Yorkshire Terrier breed, perioperative pneumonia or tracheitis, need for an additional stent placement procedure, type of tracheal collapse, and mainstem bronchial collapse or compression at the time of stent placement.
Kaplan-Meier survival curve (solid line) and associated 95% confidence intervals (dashed lines) for 75 dogs with tracheal collapse treated by endoluminal tracheal stent placement.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Kaplan-Meier survival curve (solid line) and associated 95% confidence intervals (dashed lines) for 75 dogs with tracheal collapse treated by endoluminal tracheal stent placement.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Kaplan-Meier survival curve (solid line) and associated 95% confidence intervals (dashed lines) for 75 dogs with tracheal collapse treated by endoluminal tracheal stent placement.
Citation: Journal of the American Veterinary Medical Association 254, 3; 10.2460/javma.254.3.380
Results of Kaplan-Meier analysis to identify variables associated with survival time after initial stent placement for the dogs of Table 1.
| Variable | No. of dogs | No. of deaths | Median (IQR) survival time (d) | P value |
|---|---|---|---|---|
| Age (y) | < 0.001 | |||
| < 7.7 | 37 | 19 | 1,567 (1,005–2,050) | |
| ≥ 7.7 | 38 | 26 | 572 (107–848) | |
| Sex | 0.02 | |||
| Male | 41 | 21 | 1,270 (572–2,050) | |
| Female | 34 | 24 | 728 (193–1,183) | |
| Breed | 0.07 | |||
| Yorkshire Terrier | 50 | 29 | 1,132 (598–2,050) | |
| All other dogs | 25 | 16 | 572 (155–1,005) | |
| Type of tracheal collapse | 0.11 | |||
| TTC | 37 | 23 | 776 (107–1,567) | |
| MTC | 38 | 22 | 1,132 (598–2,050) | |
| Pneumonia or tracheal infection† | 0.34 | |||
| Detected | 39 | 21 | 764 (107–NA) | |
| Not detected | 36 | 24 | 1,087 (685–1,805) | |
| Bronchial collapse or compression‡ | 0.85 | |||
| Detected | 43 | 24 | 1,005 (209–2,050) | |
| Not detected | 18 | 13 | 868 (602–1,488) | |
| History of cardiac abnormality§ | 0.03 | |||
| Present | 21 | 17 | 728 (155–1,183) | |
| Not present | 64 | 28 | 1,132 (517–2,050) | |
| Major postoperative complication | ||||
| Stent fracture | 13 | 8 | 776 (517–839) | 0.02‖ |
| Tissue ingrowth | 12 | 6 | 1,805 (724–NA) | 0.96¶ |
| Progressive tracheal collapse | 7 | 5 | 1,488 (848–2,050) | |
| No. of stent placement procedures | 0.33 | |||
| 1 | 42 | 25 | 868 (184–1,920) | |
| ≥ 2 | 33 | 20 | 1.005 (598–1,805) | |
P values represent the results of the log-rank test.
Detected perioperatively.
Detected at or prior to the time of stent placement.
History of cardiac abnormality could have included cardiomegaly and heart murmur.
P value from comparison of dogs with stent fracture with dogs with tissue ingrowth or progressive tracheal collapse.
P value for comparison between dogs with tissue ingrowth and dogs with progressive tracheal collapse.
IQR = Interquartile (25th to 75th percentile) range. NA = Not available because of lack of deaths reported.
Results of Cox proportional hazards analysis indicated that as age increased, so did the risk of death over the study period (hazard ratio, 1.30; 95% confidence interval, 1.17 to 1.45; P < 0.001). In separate analyses, age (as a continuous variable) was analyzed in the presence of other potential confounding effects (Table 4). Even in the presence of a given confounding effect, age continued to be significant.
Results of Cox proportional hazard modeling to identify variables associated with survival over the study period for the dogs of Table 1, controlling for age.
| Model No. | Variables | Hazard ratio (95% confdence interval) | P value |
|---|---|---|---|
| 1 | Age (y) | 1.28 (1.15–1.44) | < 0.001 |
| Cardiac abnormality (yes vs no) | 1.29 (0.69–2.42) | 0.43 | |
| 2 | Age (y) | 1.30 (1.16–1.45) | < 0.001 |
| Stent fracture requiring a second stent placement procedure (yes vs no) | 0.95 (0.43–2.10) | 0.89 | |
| 3 | Age (y) | 1.30 (1.15–1.46) | < 0.001 |
| MTC (vs TTC) | 0.99 (0.51–1.93) | 0.99 |
A significant (P < 0.05) hazard ratio > 1 indicates an increased risk of death relative to the referent category (categorical variables) or the increase in the risk of death for each 1-year increase in age.
Discussion
During the nearly 6-year span of the present study, 157 dogs with tracheal collapse were evaluated by clinicians of the interventional radiology service of the study hospital. This caseload represented approximately 1 new case every 2 weeks and suggested that this condition continues to be a substantial concern. Although 7 breeds or types were represented in the 75 dogs that were included in the study, Yorkshire Terriers and Pomeranians accounted for > 80% of patients. Age at the time of stent placement ranged between 2 and 15 years, supporting previous evidence that this condition can manifest at any time with a variable rate of progression.1–4
The MTC and TTC types of tracheal collapse were nearly equally distributed among the dogs. As suspected, dogs with MTC differed from those with TTC in some signalment characteristics (mean age and proportions of Yorkshire Terriers), but not in other characteristics or in outcomes. Although younger dogs with tracheal collapse (regardless of type) lived longer and were more likely to have tracheal malformation than older dogs, the difference in MSTs between dogs with MTC (1,132 days) and TTC (776 days) was not significant (P = 0.11) in this population of dogs. These data suggested that dogs with MTC should not be perceived as having more severe or progressive disease than dogs with TTC, as might be implied by the grade of tracheal collapse in these dogs (grade IV; nearly 100% luminal occlusion); however, the authors now consider this variant as a possible different form of TC, rather than an extension or progression (grade IV) of the historically classified TC grading scheme.1 The term malformation can be defined as an abnormality of shape or form22 and was chosen to describe dogs with W-shaped malformation of the tracheal cartilages for that reason and not to suggest that the condition is congenital, which has not been established.
Although most dogs in the present study were already being managed medically prior to initial stent placement (median duration, 207 days), 15% were not. The authors prefer to maximize medical management of clinical signs associated with tracheal collapse before resorting to stent placement; however, this is not always possible when dogs are in respiratory crisis. Indeed, 52% of dogs included in the present study were admitted to the hospital through the emergency service rather than through an elective appointment with the interventional radiology service. The emergent status of many of these dogs was further supported by the mean clinical severity scores for goose-honking or raspy breathing and dyspnea (7/10 for both), which suggested fairly severe respiratory clinical signs. The low mean clinical severity score for non–goose-honk coughing (2/10) was expected because this is more often a sign of lower airway and cardiac disease than of tracheal collapse. Common comorbidities identified in dogs for which thoracic radiography was performed prior to stent placement included cardiomegaly (33%) and hepatomegaly (33%), as previously reported.2–4 Rib fractures, identified in 8% of study dogs, were further evidence of the severity of respiratory problems that some dogs had.
Although the procedure for fluoroscopic stent placement was fairly uniform over the study period, the nature of the ancillary diagnostic procedures performed changed. Toward the latter half of the study period, tracheoscopy was performed for all dogs before and after stent placement as well as endotracheal lavage after stent placement to collect samples for bacterial culture and antimicrobial susceptibility testing. These additional techniques helped confirm appropriate stent size and position and allowed detection of a surprisingly high number of tracheal infections or colonizations after the first stent placement procedure (79% of tested dogs). Authors of a previous report23 suggested bacterial infections do not contribute substantially to clinical signs in dogs with tracheal collapse, citing the presence of multiple bacterial species but little cytologic evidence of active infection (ie, few neutrophils and monocytes). In that report,23 Pseudomonas spp, Enterobacter spp, and other bacteria were commonly recovered from endotracheal lavage samples from dogs with tracheal collapse. On the contrary, in the present study, typically uniform populations of bacteria were identified, and cytologic evaluation often revealed neutrophilic and monocytic inflammation suggestive of a possible underlying infectious process. The difference in findings between studies may be attributable to different severities of tracheal collapse in the included dogs.
Mainstem bronchial collapse or compression was noted in 70% of dogs at the time of stent placement in the present study, similar to previous findings.11,24 However, no difference in MST was identified between dogs with versus without this abnormality at the time of stent placement. We hypothesized that this would be the case because in our experience, most dogs with tracheal collapse have identifiable mainstem bronchial collapse or compression on tracheobronchoscopy; however, most of the clinical signs are associated with inspiratory dyspnea (as expected most commonly with extrathoracic lesions) rather than expiratory dyspnea (as expected most commonly with intrathoracic lesions).
Median duration of the combined endoscopy, stent placement, and endotracheal lavage procedures was 30 minutes and increased to only 35 minutes when ancillary surgical procedures were included. Although laryngeal paralysis can occur in up to 25% of dogs with tracheal collapse,1–4 no laryngeal paralysis was identified in the dogs of the present study. However, 24% of dogs had an elongated soft palate requiring partial staphylectomy and 8% had epiglottic retroversion necessitating epiglottopexy. We believe such concurrent upper airway obstructions contributed substantially to the dogs’ airway crises and need to be more carefully assessed when treating dogs with tracheal collapse in the future.
The nearly 2-year median follow-up time for dogs in the present study represented the longest period reported for patients receiving endoluminal tracheal stents to manage tracheal collapse.11,14,19,21 Improvement in clinical severity scores for goose-honking or raspy breathing and dyspnea by the final recheck examination was achieved for 89% and 84% of dogs, respectively, consistent with our hypothesis. The long-term need for continued corticosteroid and antitussive treatment was also anticipated, given the continued presence of a foreign object (stent) within the trachea.10,11
Survival and perioperative complication rates in the study reported here were similar to those previously reported.11,14,19,21 Recorded complications were similar to those previously reported, including stent fracture, tissue ingrowth, progressive tracheal collapse, and infection. In addition, although stent fracture and tissue ingrowth were more likely to be noted during the longer-term follow-up periods, all types of noted complications occurred in all follow-up periods, suggesting the importance of regular recheck examinations. Although stent migration and stent shortening have reportedly occurred in up to 37%12 and 27%14,21 of patients, respectively, substantial complications of this nature were not recorded for the study dogs. A subgroup of the 75 study dogs was previously evaluated for short-, immediate-, and long-term stent shortening and migration,25 and stent shortening of approximately 10% and no evidence of significant stent migration was identified in 54 dogs over a mean follow-up period of 452 days.
Progressive tracheal collapse beyond the stent was an important reason that dogs in the present study required an additional stent placement procedure, particularly dogs with MTC. This collapse was not, however, secondary to stent shortening as previously reported. Rather, initial stents placed earlier in the study period for dogs with MTC were positioned solely across the malformed segment of trachea at the thoracic inlet, whereas such stents for dogs with TTC were placed along the entire length of the trachea. Later in the study period, stents in dogs with MTC were placed closer to the full length of the trachea, as in dogs with TTC. Although stents did not differ significantly in original length between the first and second half of the patients, larger-diameter stents were used for the second half, particularly for dogs with MTC, and such stents will ultimately increase in length when expanded to the same tracheal diameters owing to their woven and foreshortening nature. Increasing the stent diameter improved tracheal wall contact while also providing coverage of a longer length of the trachea.
Stent fracture remains one of the major long-term complications associated with endoluminal tracheal stents. Although stent fracture was the most common reason for an additional stent placement procedure for dogs with TTC and tissue ingrowth was the most common reason for dogs with MTC, the proportion of dogs with stent fractures was similar in both groups. In the authors’ experience, fractures generally occur along the dorsal, intrathoracic portion of an endoluminal tracheal stent, possibly because of material fatigue and plastic deformation in this location where there is minimal cartilage support and restricted stent expansion. Improvements have since been made in some commercially available SEMSs to address those concerns.
Tissue ingrowth can occur when the gaps (or gutters) between the stent and tracheal wall prevent full epithelialization of the endoluminal stent and contribute to mucus pooling and likely subsequent infection. Such ingrowth is not attributable to granulation tissue per se,14,18 as supported by the finding that it was successfully medically managed in 50% of affected dogs in the present study. Without frequent recheck examinations, tissue ingrowth may go undiagnosed until a crisis occurs necessitating intubation, stent replacement, or other respiratory interventions. Our experience suggests that gaps between the stent and tracheal mucosa should be avoided whenever possible.
Tracheal infections and pneumonia were included as a single complication in the study reported here because it can be difficult to differentiate between the two. The conditions were categorized as minor complications when managed medically or major complications if considered a cause of death. Although 2 of 5 perioperative deaths were considered due to pneumonia, the pneumonia was likely present prior to stent placement. Overall, pneumonia or tracheal infection was the most common complication (57%); however, 12% of dogs had a history of pneumonia prior to stent placement, and 79% had positive results of bacterial culture at the time of stent placement. The similar proportion of dogs with positive culture results after hospital discharge following stent placement (78%) may have suggested that the presence of a stent does not increase the risk of subsequent infection.
The only factors identified as associated with survival time in the present study were male sex and age, with male dogs and younger dogs having significantly longer MSTs than female dogs and older dogs. Notably, no association with survival time was identified for presence of mainstem bronchial collapse at the time of stent placement, Yorkshire Terrier breed, additional stent placement procedures required, and other evaluated variables. Although these findings would need to be confirmed, this important information could be helpful to provide to clients considering a stent placement procedure for their dogs. In addition, although approximately 19% of dogs required an additional stent following stent fracture, only one of these dogs (n = 13) went on to develop a second stent fracture. However, one-quarter (3/12) of dogs that received a second stent because of tissue ingrowth required an additional stent afterward, suggesting vigilant follow-up is required for this subset of patients and perhaps additional, more aggressive medical management.
Findings of the present study suggested that endoluminal tracheal stent placement can provide a low-mortality option for management of tracheal collapse even in severely affected dogs; however, complications are possible, particularly pneumonia or tracheal infections, stent fractures, and tissue ingrowth. Long-term improvement in clinical severity scores for goose-honking or raspy breathing and dyspnea was achieved for 89% and 84% of dogs, respectively, and fairly long survival times were possible for certain subsets, primarily younger dogs. Coughing (non–goose-honking) should not be expected to improve as much as the other clinical severity scores, given that it worsened or improved in an approximately equal percentage of patients. Dogs with MTC and TTC in this study population may have differed in some factors and outcomes, but they did not differ significantly in survival times or rates.
Acknowledgments
This study was completed in the Interventional Radiology Department at the Animal Medical Center, New York, NY.
Funded by the Egg Roll Fund.
Dr. Weisse is a consultant for, and minority partner of, Infiniti Medical LLC.
Preliminary findings presented as a poster at the American College of Veterinary Surgeons Surgery Summit, San Diego, October 2014.
ABBREVIATIONS
| MST | Median survival time |
| MTC | Malformation type of tracheal collapse |
| SEMS | Self-expanding metallic stent |
| TTC | Traditional type of tracheal collapse |
Footnotes
Weasel wire, Infiniti Medical LLC, Menlo Park, Calif.
Measuring catheter, Infiniti Medical LLC, Menlo Park, Calif.
Vet Stent-Trachea, Infiniti Medical LLC, Menlo Park, Calif.
Wallstent endovascular stent, Boston Scientific, Natick, Mass.
SAS, version 9.4, SAS Institute Inc, Cary, NC.
References
1. Tanger CH, Hobson HP. A retrospective study of 20 surgically managed cases of collapsed trachea. Vet Surg 1982;11:146–149.
2. Sura PA, Durant AM. Trachea and bronchi. In: Tobias KM, Johnston SA, eds. Veterinary surgery: small animal. St Louis: Elsevier Saunders, 2012;1746–1750.
3. Payne JD, Mehler SJ, Weisse C. Tracheal collapse. Compend Contin Educ Pract Vet 2006;28:373–382.
4. Buback JL, Boothe HW, Hobson HP. Surgical treatment of tracheal collapse in dogs: 90 cases (1983–1993). J Am Vet Med Assoc 1996;208:380–384.
5. Weisse C. Intraluminal tracheal stenting. In: Weisse C, Berent A, eds. Veterinary image-guided interventions. Singapore: Wiley Blackwell, 2015;73–82.
6. Hobson HP. Total ring prosthesis for the surgical correction of collapsed trachea. J Am Anim Hosp Assoc 1976;12:822–828.
7. Kirby BM, Bjorling DE, Rankin JHG, et al. The effects of surgical isolation and application of polypropylene spiral prostheses on tracheal blood flow. Vet Surg 1991;20:49–54.
8. White RN. Unilateral arytenoid lateralisation and extraluminal polypropylene ring prostheses for correction of tracheal collapse in the dog. J Small Anim Pract 1995;36:151–158.
9. Becker WM, Beal M, Stanley BJ, et al. Survival after surgery for tracheal collapse and the effect of intra-thoracic collapse on survival. Vet Surg 2012;41:501–506.
10. Chisnell HK, Pardo AD. Long-term outcome, complications and disease progression in 23 dogs after placement of tracheal ring prostheses for treatment of extrathoracic tracheal collapse. Vet Surg 2015;44:103–113.
11. Tinga S, Thieman Mankin KM, Peycke LE, et al. Comparison of outcome after use of extra-luminal rings and intra-luminal stents for treatment of tracheal collapse in dogs. Vet Surg 2015;44:858–865.
12. Radlinsky MG, Fossum TW, Walker MA, et al. Evaluation of the Palmaz stent in the trachea and mainstem bronchi of normal dogs. Vet Surg 1997;26:99–107.
13. Gellasch KL, DáCosta Gómez T, McAnulty JF, et al. Use of intraluminal nitinol stents in the treatment of tracheal collapse in a dog. J Am Vet Med Assoc 2002;221:1719–1723.
14. Moritz A, Schneider M, Bauer N. Management of advanced tracheal collapse in dogs using intraluminal self-expanding biliary wallstents. J Vet Intern Med 2004;18:31–42.
15. Mittleman E, Weisse C, Mehler SJ, et al. Fracture of an endoluminal nitinol stent used in the treatment of tracheal collapse in a dog. J Am Vet Med Assoc 2004;225:1217–1221.
16. Ouellet M, Dunn ME, Lussier B, et al. Noninvasive correction of a fractured endoluminal nitinol tracheal stent in a dog. J Am Anim Hosp Assoc 2006;42:467–471.
17. Woo HM, Kim MJ, Lee SG, et al. Intraluminal tracheal stent fracture in a Yorkshire Terrier. Can Vet J 2007;48:1063–1066.
18. Brown SA, Williams JE, Saylor DK. Endotracheal stent granulation stenosis resolution after colchicine therapy in a dog. J Vet Intern Med 2008;22:1052–1055.
19. Sura PA, Krahwinkel DJ. Self-expanding nitinol stents for the treatment of tracheal collapse in dogs: 12 cases (2001–2004). J Am Vet Med Assoc 2008;232:228–236.
20. Sun F, Uson J, Ezquerra J, et al. Endotracheal stenting therapy in dogs with tracheal collapse. Vet J 2008;175:186–193.
21. Durant AM, Sura P, Rohrbach B, et al. Use of nitinol stents for end-stage tracheal collapse in dogs. Vet Surg 2012;41:807–817.
22. English Oxford Living Dictionaries. Malformation. Available at: en.oxforddictionaries.com/definition/malformation. Accessed Mar 23, 2017.
23. Johnson LR, Fales WH. Clinical and microbiologic findings in dogs with bronchoscopically diagnosed tracheal collapse: 37 cases (1990–1995). J Am Vet Med Assoc 2001;219:1247–1250.
24. Johnson LR, Singh MK, Pollard RE. Agreement among radiographs, fluoroscopy, and bronchoscopy in documentation of airway collapse in dogs. J Vet Intern Med 2015;29:1619–1626.
25. Raske M, Weisse C, Berent A, et al. Immediate, short-term, and long-term changes in tracheal stent diameter, length, and positioning following placement in dogs with tracheal collapse syndrome: 53 cases. J Vet Intern Med 2018;32:782–791.
