Nasopharyngeal stenosis and INP are uncommon disorders in dogs and cats that narrow or occlude the nasopharynx, diminishing or obstructing airflow through the nasal passages. These disorders may be congenital; acquired secondary to an infectious, inflammatory, caustic (eg, aspiration rhinitis), or traumatic etiology1–6; a result of benign or malignant masses1; or a result of scarring following nasopharyngeal surgery.4,6–8
Both NPS and INP can result in stertorous nasal breathing, dyspnea, open-mouth breathing, gagging, repeated swallowing, sneezing, and chronic serous or mucopurulent nasal discharge.1,3,6–14 A definitive diagnosis can be made by means of retropharyngoscopy, CT of the nasopharynx, or both.1,6,7,9–15
Reported treatment options for NPS include surgical resection, laser ablation, mechanical dilatation with a Kelly forceps or balloon, and stenting with a metallic stent or silicone tubing.3,6–8,10–14,16–18 In numerous reports,8,10,11,15,17 balloon dilatation has been reported to result in an immediate resolution of clinical signs, but with high recurrence rates. A recent study18 reported successful outcomes following placement of silicone tubing in the nasopharynx of cats for 3 to 4 weeks after dilatation of the stenosis.
Placement of UMSs and CMSs has been described for the treatment of NPS and INP in small numbers of dogs and cats.6,8,12,13 There were no immediate complications associated with stent placement in these patients, and immediate relief of stertorous breathing was achieved. Long-term complications included tissue ingrowth, oronasal fistula development, granulation tissue development, hair ball entrapment, stent compression, dysphagia, and chronic infection.6,8,12 Most of these complications were relieved with additional interventions, and long-term outcomes were considered good by pet owners and their veterinarians.
To our knowledge, all previous reports on dogs and cats with NPS and INP have been small case series, with no studies comparing results for dogs versus cats, for NPS versus INP, or among the various treatment modalities. The objective of the study reported here, therefore, was to determine outcomes for a larger group of dogs and cats with benign NSP or INP that underwent balloon dilatation or stent placement. We hypothesized that single balloon dilatation procedures would typically not be successful long term and that stent placement would be associated with a higher success rate (ie, maintenance of a patent nasopharynx) but also with a higher complication rate.
Materials and Methods
Case selection criteria
Medical records of the Animal Medical Center for June 2005 to October 2013 were searched to identify dogs and cats with benign NPS or INP that was treated by means of balloon dilatation or placement of a CMS or UMS. Patients were excluded if < 6 months of follow-up information after treatment of NPS or INP was available, except that patients that died or were euthanized < 6 months after treatment because of complications associated with NPS or INP treatment were included. Six patients previously reported6 were included because an additional 5 years of follow-up information was available, providing a better assessment of long-term complications.
Historical and laboratory data
For patients included in the study, information obtained from the medical records consisted of signalment, age at diagnosis, concurrent illnesses, and whether there was any history of upper respiratory tract infections, previous anesthetic events, or episodes of chronic rhinitis. In all patients, a CBC and serum biochemistry profile were performed prior to treatment. Retropharyngoscopy was performed in all patients to confirm the diagnosis; in some patients, CT or MRI of the nasopharynx was also performed prior to treatment. In patients in which a stent was placed, fluoroscopic nasopharyngography with contrast medium was performed to further define the lesion and determine the appropriate stent size.
Balloon dilatation and stent placement
Patients were anesthetized, underwent an oral and laryngneal examination, and were then endotracheally intubated. Patients were then positioned in right lateral recumbency, and retropharyngoscopy was performed with a flexible gastroscopea (for the 6 previously reported patients6) or flexible bronchoscopeb (for all remaining patients) to evaluate the nasopharynx and choanae, determine whether NPS or INP was present, and localize the lesion to the caudal, middle, or rostral third of the nasopharynx. Results of nasopharyngography, CT, and MRI were used to assist in lesion localization when available.
For patients with NPS, a 0.035-inch hydrophilic, angled-tipped guidewirec was advanced antegrade into the nares and ventral nasal meatus and through the stenotic region under endoscopic guidance. When fluoroscopy was available, a multifenestrated marker catheterd was passed over the guidewire and advanced until fenestrations were positioned on either side of the stenotic region. Care was taken to ensure the endotracheal tube cuff was inflated appropriately to avoid contrast medium aspiration, and contrast nasopharyngography was performed with a 1:1 mixture of iohexole and saline (0.9% NaCl) solution to further evaluate the length, location, and diameter of the stenotic region (Figure 1).
Images of a 1-year-old spayed female Chinese Crested dog with NPS; the patient is in right lateral recumbency. A—Endoscopic image of the NPS. B—Fluoroscopic image showing a guidewire being passed through the NPS (white arrow). The retroflexed endoscope can also be seen (white arrowhead). C—Digital-subtraction fluoroscopic image obtained during contrast nasopharyngography. A marker catheter (white arrow) has been placed over the guidewire, and the guidewire has been removed. A 1:1 mixture of iohexol and saline (0.9% NaCl) solution was then injected, revealing the NPS (black arrow). The retroflexed endoscope can also be seen in this image (white arrowhead).
Citation: Journal of the American Veterinary Medical Association 253, 10; 10.2460/javma.253.10.1300
For patients with INP, the lesion was pierced as previously described6 (Figure 2). In brief, a 6F or 8F vascular access sheathf was advanced over a guidewire in an antegrade direction to the level of the lesion under fluoroscopic guidance. An 18-gauge, 15-cm-long trocar needleg was then inserted into the sheath and, once visualized endoscopically and fluoroscopically, was used to puncture the membrane from the rostral aspect. The guidewire was then inserted antegrade through the trocar needle and advanced to the level of the middle or distal portion of the esophagus to ensure that the stiffest part of the guidewire was in the area of the lesion, and that access was maintained. Vascular dilators and a balloon dilatation catheter were used to dilate the guidewire tract, and contrast nasopharyngography was then performed as described for patients with NPS.
Endoscopic images of a 2.5-year-old spayed female mixed-breed dog with INP; the patient is in right lateral recumbency. A—Image showing the INP. B—A vascular access sheath was placed antegrade from the nares to the imperforate membrane, and an 18-gauge, 15-cm-long trocar needle (asterisk) inserted into the vascular access sheath was used to puncture the membrane. A 0.035-inch guidewire (arrow) was then advanced through the trocar and into the esophagus.
Citation: Journal of the American Veterinary Medical Association 253, 10; 10.2460/javma.253.10.1300
In patients that underwent balloon dilatation alone, the procedure was performed by 1 of 2 authors (ACB or DP). The marker catheter used for contrast nasopharyngography was withdrawn, and an appropriately sized balloon catheterh was chosen on the basis of measurements obtained by means of fluoroscopy, endoscopy, or CT (Figure 3). Balloon dilatation was performed as described.17 The balloon catheter was passed over the guidewire through the stenotic area. A balloon insufflation devicei was then used to inflate the balloon to the recommended maximal inflation pressure with a 1:1 mixture of iohexol and saline solution. The balloon was held in place until effacement of the stenosis was observed endoscopically or fluoroscopically, typically within 30 to 60 seconds.
Endoscopic and fluoroscopic images of the same dog as in Figure 1; the patient remains in right lateral recumbency. The marker catheter has been withdrawn, and an appropriately sized balloon catheter was passed over the guidewire through the stenotic region. A—Endoscopic view of the insufflated balloon through the stenosis (black asterisk). B—Fluoroscopic image of the balloon (white asterisk) following insufflation with a 1:1 mixture of iohexol and saline solution. The flexible endoscope can also be seen (arrowhead).
Citation: Journal of the American Veterinary Medical Association 253, 10; 10.2460/javma.253.10.1300
In patients in which metallic stents were placed, the procedure was performed by 1 or 2 authors (ACB and CW). Stents were placed with endoscopic and fluoroscopic guidance as described6 (Figure 4). Briefly, the stenosis was balloon dilated, and an appropriate stent size was chosen. A UMSj–l (most were BEMSs) or CMSj,m,n (BEMS or SEMS) was then passed with a stent delivery system over the guidewire and through the stenosis, under endoscopic and fluoroscopic guidance. Once the stent was properly positioned, the endoscope was backed out of the nasopharynx slightly to visualize the caudal end of the stenosis and leading edge of the stent. The stent was then deployed, and proper location was determined endoscopically and fluoroscopically.
Endoscopic and fluoroscopic images of a 5-year-old spayed female domestic shorthair cat with NPS; the patient is in right lateral recumbency. A—Endoscopic image of a UMS in place over a guidewire (white asterisk). B—Fluoroscopic image after a UMS (arrow) was placed through the NPS; the guidewire remains in place (black asterisk). The flexible endoscope can be seen (arrowhead).
Citation: Journal of the American Veterinary Medical Association 253, 10; 10.2460/javma.253.10.1300
Following balloon dilatation or stent placement, the nose was flushed with sterile saline solution and suctioned with the endoscope still positioned in the nasopharynx. In some cases, triamcinolone (0.2 mg/kg [0.09 mg/lb]) was injected submucosally or 0.1% mitomycin C was applied topically, depending on clinician preference. Depending on the location and anatomy of the stenotic region or whether stent migration was considered likely, the stent was sutured in place with a tacking suture from the caudal aspect of the soft palate through the interstices of the stent. Bupivacaine (0.6 mg/kg [0.27 mg/lb] in dogs; 0.3 to 0.6 mg/kg [0.14 to 0.27 mg/lb] in cats) was applied topically after completion of the procedure, and patients were treated with buprenorphine (0.1 mg/kg [0.045 mg/lb], IV) as needed for discomfort after the procedure.
Postsurgical management
At discharge, owners were provided with an analgesic to be administered for 1 to 3 days if signs of discomfort were noted. This was typically buprenorphine (0.01 mg/kg [0.0045 mg/lb], transmucosally, q 8 to 12 h) or tramadol (2 to 4 mg/kg [0.9 to 1.8 mg/lb], PO, q 8 to 12 h). Amoxicillin–clavulanic acid (15 mg/kg [6.8 mg/lb], PO, q 12 h, for 14 to 28 days) and prednisone (0.5 to 1.0 mg/kg [0.23 to 0.45 mg/lb], PO, q 12 h, for 14 days and then a tapering dosage over 2 weeks) were also prescribed.
Follow-up
Patients were reassessed via physical examination either 6 to 12 weeks (for the 6 previously reported patients6) or 4 weeks (for all remaining patients) after the procedure. All additional procedures performed and complications (ie, tissue regrowth after balloon dilatation, tissue ingrowth through the stent, chronic infection, oronasal fistula formation, stent fracture, stent bending, stent migration, exaggerated swallowing, and stent removal) were recorded. All clients were contacted via telephone intermittently thereafter. Outcome was defined as success or failure on the basis of clinical improvement (with a successful outcome defined as normal airflow through the nasopharynx, lack of inspiratory stridor, decreased nasal discharge, and normal attitude, appetite, and activity level), the complications encountered, and the owner's level of satisfaction (ie, happiness with pet's quality of life, attitude, appetite, and activity level).
Statistical analysis
The Kolmogorov-Smirnoff test was used to determine whether baseline descriptive data were normally distributed. Normally distributed data were summarized as mean and SD; nonnormally distributed data were summarized as median and range. Continuous baseline variables were compared between groups by means of ANOVA (parametric data) or the Wilcoxon test (nonparametric data). Categorical variables were compared between groups with χ2 or Fisher exact tests. All analyses were performed with commercially available statistical software.o Values of P < 0.05 were considered significant. Given the exploratory nature of the study, no adjustments were made for multiple analyses.
Results
Fifteen dogs and 31 cats met the criteria for inclusion in the study. Information for 3 of the dogs and 3 of the cats has been published previously,6 but these patients were included in the study because additional follow-up information had been obtained. There were 2 mixed-breed dogs and 1 each of English Springer Spaniel, Old English Sheepdog, West Highland White Terrier, Pug, Shih Tzu, Chinese Shar Pei, Chihuahua, Irish Setter, Viszla, Chinese Crested, Beagle, Dachshund, and Miniature Poodle. There were 23 domestic shorthair cats, 2 domestic longhair cats, 2 Ocicats, and 1 each of Siamese, Devon Rex, Burmese, and Tonkinese. Median age at the time of the procedure was 13.2 months for the dogs (range, 6 to 198 months) and 36 months for the cats (range, 3 to 180 months). Median follow-up time was 24 months (range, 2 to 109 months). One dog was euthanized because of a lack of improvement 2 months after stent placement. No patients were excluded.
Continuous stertorous breathing was identified in all 46 patients. The suspected cause of NPS or INP was chronic rhinitis in 8 patients (1 dog and 7 cats), postanesthetic aspiration rhinitis in 7 (6 dogs and 1 cat), upper respiratory tract infection in 4 (all cats), suspected congenital webbing in 3 (1 dog and 2 cats), postsurgical or post-traumatic scarring in 2 (both dogs), and aspiration rhinitis not associated with anesthesia in 1 (a dog). The cause of NPS or INP was not determined in 21 patients (4 dogs and 17 cats).
Preanesthetic serum biochemical profiles were evaluated in all patients. Mild hyperglycemia was noted in 4 patients, hypercholesterolemia was noted in 3, mild increases in serum alanine transaminase and creatine kinase activities were noted in 2, and high serum total protein and albumin concentrations were noted in 2. Complete blood counts were also evaluated in all patients, and neutrophilia was detected in 5 patients, monocytosis was detected in 4, and mild nonregenerative anemia was detected in 2. Computed tomography was performed in 19 patients (7 dogs and 12 cats) and confirmed the diagnosis of NPS or INP in all 19. Magnetic resonance imaging was performed in 3 cats and confirmed the diagnosis of NPS in 2. Endoscopic evaluation confirmed the presence of NPS or INP in all patients. Nasopharyngography was performed in all patients prior to balloon dilatation or stent placement.
Eight of the 15 dogs had INP, and 7 had NPS. Lesions were located in the rostral third of the nasopharynx in 13 dogs and in the middle third in 1 (in the remaining dog, lesion location was not reported). Twenty-eight of the 31 (90%) cats had NPS, and 3 (10%) had INP. Lesions were located in the caudal third of the nasopharynx in 22 (71%) cats, in the rostral third in 5 (16%), and in the middle third in 4 (13%).
Balloon dilatation
Twenty-seven patients (5 dogs and 22 cats) underwent balloon dilatation. Results were considered successful in 11 (0 dogs and 11 cats) of the 27 (41%) patients, as determined by lack of tissue regrowth and absence of clinical signs after 1 or more balloon dilatation procedures. Outcome of balloon dilatation (success vs failure) was not significantly (P = 0.054) associated with species (dog vs cat). Three of the 11 cats in which balloon dilatation was successful required 2 dilatation procedures. In the remaining 8 cats, balloon dilatation was successful after a single procedure.
Median number of balloon dilatation procedures was 1 (range, 1 to 3). Seven of 21 patients in which a single balloon dilatation procedure was performed had a successful outcome, compared with 4 of 5 patients in which 2 balloon dilatation procedures were performed. Only 1 patient had 3 balloon dilatation procedures performed; outcome was considered unsuccessful in this patient. Number of balloon dilatation procedures was significantly (P = 0.045) associated with outcome (success vs failure). Repeated balloon dilatation procedures were performed a median of 6 weeks apart (range, 2 to 49 weeks). Balloon diameter ranged from 6 to 18 mm (median, 9 mm) in cats and from 10 to 12 mm (median, 11 mm) in dogs. Median balloon diameter for patients in which balloon dilatation was unsuccessful (8.5 mm; range, 6 to 12 mm) was not significantly different from median balloon diameter for patients in which the procedure was successful (10 mm; range, 8 to 18 mm), and balloon diameter was not significantly (P = 0.098) associated with outcome. For patients in which balloon dilatation was unsuccessful, median time for tissue regrowth was 4 weeks (range, 1 to 52 weeks).
Of the 27 patients that underwent balloon dilatation, 21 had NPS and 6 had INP. Eleven of the 21 (52%) patients with NPS that underwent balloon dilatation had a successful outcome, but none of the 6 patients with INP that underwent balloon dilatation had a successful outcome. Outcome (success vs failure) was significantly (P = 0.027) associated with lesion type (NPS vs INP).
Also, of the 27 patients that underwent balloon dilatation, 8 had a lesion in the rostral third of the nasopharynx, 2 had a lesion in the middle third, and 17 had a lesion in the caudal third. Balloon dilatation was not successful in any of the patients with a lesion in the rostral or middle third of the nasopharynx, but was successful in 11 of the 17 patients with a lesion in the caudal third of the nasopharynx. Outcome (success vs failure) was significantly (P = 0.004) associated with lesion location.
Stent placement
Thirty-four patients (14 dogs and 20 cats) had a metallic stent placed, including 4 of the 5 dogs and all 11 cats in which balloon dilatation had previously been unsuccessful. No technical perioperative (< 7 days after the procedure) complications were reported after stent placement.
Uncovered metallic stents were placed in 30 patients (12 dogs and 18 cats), including 4 cats in which 2 UMSs were placed (thus, a total of 34 UMSs were placed). Outcome was classified as successful in 20 of the 30 (67%) patients in which UMSs were placed.
Covered metallic stents were placed in 11 patients (5 dogs and 6 cats), including 1 dog in which 2 CMSs were placed (thus, a total of 12 CMSs were placed). This also included 7 patients (3 dogs and 4 cats) in which CMSs were placed after loss of nasopharyngeal patency following placement of a UMS. All 11 (100%) patients in which CMSs were placed had a successful outcome.
For the cats, median stent diameter for UMSs and CMSs combined was 8 mm (range, 7 to 10 mm), and median stent length was 20 mm (range, 16 to 34 mm). For the dogs, median stent diameter for UMSs and CMSs combined was 10 mm (range, 7 to 16 mm), and median stent length was 30 mm (range, 20 to 40 mm). Median overall procedure time was 62 minutes (range, 19 to 180 minutes), but median procedure time was 58 minutes (range, 19 to 95 minutes) for patients with an NPS and 115.5 minutes (range, 47 to 180 minutes) for patients with an INP. Two of 4 patients with only CMSs, 19 of 23 (83%) patients with only UMSs, and 4 of 7 patients in which a UMS was placed followed by a CMS had a successful outcome, but use of a UMS versus a CMS was not significantly (P = 0.215) associated with outcome.
Eleven patients (3 dogs and 8 cats) had mitomycin C applied topically following the procedure, and 13 (9 dogs and 4 cats) had triamcinolone injected submucosally. Eight of the 11 patients treated with mitomycin C had tissue regrowth or ingrowth, as did 9 of the 13 patients treated with triamcinolone. Treatment with mitomycin C (yes vs no) or triamcinolone (yes vs no) was not significantly (P = 0.082 and 0.181, respectively) associated with whether tissue regrowth or ingrowth occurred. Owing to the small sample size and variety of treatments (balloon dilatation, UMS placement, and CMS placement), we could not test whether mitomycin C or triamcinolone treatment was associated with outcome of specific treatments.
Stents were sutured in place in 4 of 34 patients (12%). Two of these stents were UMSs and 2 were CMSs. Suturing of stents in place (yes vs no) was not significantly associated with whether complications occurred (P = 0.420) or with whether specific individual complications occurred (chronic infection, P = 0.157; tissue ingrowth, P = 0.436; oronasal fistula, P = 0.159; stent fracture, P = 0.512; stent migration, P = 0.291; exaggerated swallowing, P = 0.679; stent bending, P = 0.679; or stent removal, P = 0.122).
Twenty-three of the 34 (68%) patients in which stents were placed developed complications. Of the 30 patients in which UMSs were placed, 21 (70%) developed complications, including 10 (33%) with tissue ingrowth, 7 (23%) with chronic infection, 5 (17%) with stent fracture, 4 (13%) with an oronasal fistula, 4 (13%) that underwent stent removal, 3 (10%) with exaggerated swallowing, 3 (10%) with stent bending, and 2 (7%) with stent migration (some patients had > 1 complication). Of the 11 patients in which CMSs were placed, 7 (64%) developed complications, including 7 with chronic infection, 3 with an oronasal fistula, 2 that underwent stent removal, and 1 with stent migration. Because of the stent covering, none of the patients in which CMSs were placed had tissue ingrowth. In addition, none of the patients in which CMSs were placed had stent fracture, exaggerated swallowing, or stent bending.
Type of stent (UMS vs CMS) was not significantly associated with whether any complications occurred (P = 0.111) or with whether oronasal fistula (P = 0.226), stent fracture (P = 0.404), stent migration (P = 0.372), exaggerated swallowing (P = 0.455), or stent bending (P = 0.455) occurred. Type of stent was significantly associated with whether tissue ingrowth (P = 0.001) or chronic infection (P = 0.018) occurred, with tissue ingrowth more common in patients with UMSs and chronic infection more common in patients with CMSs.
Of the total 46 metallic stents that were placed (34 UMSs and 12 CMSs), 31 were BEMSs (27 UMSs and 4 CMSs placed in 11 dogs and 16 cats) and 7 were SEMSs (2 UMSs and 5 CMSs placed in 3 dogs and 4 cats). For the remaining 5 UMSs and 3 CMSs placed in 3 dogs and 3 cats, the medical record did not indicate whether a BEMS or SEMS was used. Outcome was not significantly (P = 0.688) associated with use of a BEMS versus an SEMS.
Twenty of 27 (74%) patients with a BEMS and 3 of 7 patients with an SEMS had a complication. Stent expansion type (BEMS vs SEMS) was not significantly associated with whether any complications occurred (P = 0.079) or with whether chronic infection (P = 0.730), tissue ingrowth (P = 0.284), oronasal fistula (P = 0.792), stent fracture (P = 0.145), exaggerated swallowing (P = 0.571), or stent bending (P = 0.571) occurred. However, stent expansion type was significantly (P = 0.047) associated with whether stent migration occurred, with stent migration more common with BEMSs than with SEMSs.
Species (dog vs cat) was not significantly (P = 0.058) associated with whether complications developed, but 7 of 14 (50%) dogs and 16 of 20 (80%) cats with stents had ≥ 1 complication. Species was also not significantly associated with development of any specific complication (chronic infection, P = 0.242; tissue ingrowth, P = 0.296; oronasal fistula, P = 0.101; stent fracture, P = 0.244; stent migration, P = 0.445; exaggerated swallowing, P = 0.191; and stent bending, P = 0.191). Development of complications was not significantly (P = 0.505) associated with age at the time of the procedure.
Lesion type (NPS vs INP) was not significantly (P = 0.283) associated with whether any complications occurred or with whether chronic infection (P = 0.271), tissue ingrowth (P = 0.119), oronasal fistula (P = 0.344), stent fracture (P = 0.350), stent migration (P = 0.465), exaggerated swallowing (P = 0.296), or stent bending (P = 0.296) occurred. However, all patients with exaggerated swallowing and stent bending had NPS located in the caudal third of the nasopharynx.
Although lesion location (rostral third vs middle third vs caudal third of the nasopharynx) was not significantly (P = 0.074) associated with whether any complications occurred, it was significantly associated with development of most individual complications. Six of 18 patients with a lesion in the rostral third of the nasopharynx, 5 of 5 patients with a lesion in the middle third, and 3 of 11 patients with a lesion in the caudal third developed chronic infection (P = 0.017). Four of 18 patients with a lesion in the rostral third of the nasopharynx, 4 of 5 patients with a lesion in the middle third, and 2 of 11 patients with a lesion in the caudal third developed tissue ingrowth (P = 0.031). All 3 patients with exaggerated swallowing (P = 0.037) and all 3 patients with stent bending (P = 0.037) had lesions located in the caudal third of the nasopharynx. One of 18 patients with a lesion located in the rostral third of the nasopharynx and 2 of 5 patients with a lesion located in the middle third, but none of the patients with a lesion in the caudal third of the nasopharynx, had stent migration (P = 0.029). Oronasal fistula formation (P = 0.536) and stent fracture (P = 0.289) were not significantly associated with lesion location.
Stent removal was necessary in 6 of the 34 (18%) patients, including 4 of 30 (13%) patients with UMSs and 2 of 11 patients with CMSs, because of exaggerated swallowing and gagging, oronasal fistula development, chronic infection, or stent migration. Stent removal was not significantly associated with species (dog vs cat; P = 0.161), lesion type (NPS vs INP; P = 0.3621), lesion location (rostral vs middle vs caudal third; P = 0.101), or whether a UMS (yes vs no; P = 0.122), CMS (yes vs no; P = 0.079), BEMS (yes vs no; P = 0.445), or SEMS (yes vs no; P = 0.334) was used. Two cats that underwent stent removal had a successful outcome. In one of these cats, the stent was removed after it had migrated, and respiratory signs never recurred. In the other, the first stent fractured, and a second stent was placed after the first was removed; the outcome was considered successful for this cat. The remaining 4 patients (1 dog and 3 cats) that underwent stent removal had an unsuccessful outcome.
Combined outcome
Overall, outcome was considered successful in 36 of the 46 (78%) patients, as determined by maintenance of nasopharyngeal patency, complications encountered, and owner satisfaction. Outcome was significantly (P = 0.039) associated with species, with a successful outcome in 27 of 31 (87%) cats and 9 of 15 (60%) dogs. Outcome was also significantly (P = 0.006) associated with lesion type, with a successful outcome in 31 of 35 (89%) patients with NPS and 5 of 11 (45%) with INP and in all patients with INP in which a CMS was placed. Outcome was also significantly (P = 0.013) associated with lesion location, with a successful outcome in 20 of 22 (91%) patients with lesions in the caudal third of the nasopharynx, 12 of 18 (67%) patients with lesions in the rostral third, and 3 of 5 (60%) patients with lesions in the middle third. Finally, outcome was not significantly (P = 0.079) associated with the ultimate procedure performed (balloon dilatation vs UMS placement vs CMS placement). However, because of small numbers, this result should be interpreted cautiously.
Discussion
Results of the present study suggested that NPS and INP in dogs and cats can be successfully managed with balloon dilatation or stent placement, with 36 of 46 (78%) patients in the present study having a successful outcome.
Although outcome (success vs failure) in the present study was not found to be significantly associated with the ultimate procedure performed, the success rate following balloon dilatation alone was only 30% (8/27) after a single dilatation procedure and only 41% (11/27) after multiple dilatation procedures. That said, none of the patients in the present study developed perioperative complications associated with the procedure. Previous reports15,17 described presumptive vagus nerve–mediated bradycardia necessitating atropine administration in a dog and a cat undergoing balloon dilatation, but this complication was not encountered in the present study. For patients in the present study in which balloon dilatation was unsuccessful, median recurrence time was 4 weeks, which was similar to times reported previously.10,13,15,17
We did not find a significant association between species (dog vs cat) and outcome of balloon dilatation in the present study; however, the P value (0.054) was close to our cutoff for significance, and none of the 5 dogs in which balloon dilatation was performed had a successful outcome. Outcome of balloon dilatation was significantly associated with both lesion type (NPS vs INP) and lesion location (rostral vs middle vs caudal third of the nasopharynx), with outcome less likely to be successful for patients with INP or with a lesion in the rostral or middle third of the nasopharynx. Dogs more commonly had INP (8/15) and rostral lesions (13/15), whereas cats more commonly had NPS (28/31) and caudal lesions (22/31). Our findings may suggest that dogs have more aggressive forms of these conditions and a greater propensity for tissue regrowth, but may also suggest a different pathophysiology between locations or species.
In the present study, 3 cats required a second balloon dilatation procedure for a successful outcome, and the number of balloon dilatation procedures was significantly associated with outcome. This may have suggested that with a larger sample size and more commonly repeated dilatation procedures, the success rate for this procedure might be found to be higher. However, in our clinical experience, the financial burden of multiple dilatation procedures, compared with the cost of placing a stent, has resulted in many owners declining more than 1 or 2 balloon dilatation procedures. Importantly, most cats in the present study had been referred to our tertiary referral practice for stenting after balloon dilatation performed by other specialists had failed, and this likely biased the population toward more severely affected cases.
Currently, procedures for balloon dilatation and stent placement in dogs and cats have not been standardized. Balloon diameter was not standardized for patients in the present study, but median diameter for patients in which balloon dilatation was unsuccessful (8.5 mm) was not significantly different from median diameter for patients in which the procedure was successful (10 mm). Also, whether fluoroscopic guidance was used in addition to endoscopy for balloon dilatation was not standardized in the present study, although most referring specialists who had performed a balloon dilatation prior to referral to our institution did not use concurrent fluoroscopy. Clinically, we believe that using fluoroscopy in addition to endoscopy when performing balloon dilatation may improve the success rate for the procedure because expansion of the stenosis can be visualized to ensure the proper balloon diameter was chosen. In contrast, balloon diameter cannot be assessed when monitoring balloon dilatation with endoscopy alone, because only the mucosa is visible endoscopically. Objective evaluations of balloon diameter and of using endoscopy versus endoscopy and fluoroscopy for balloon dilatation should be performed in the future.
In the present study, there were 46 stents placed in 34 patients. Median stent size was 8 × 20 mm in cats and 10 × 30 mm in dogs. Stents were either covered or uncovered and either balloon-expandable or self-expandable; however, most of the stents were uncovered BEMSs. The type of stent used was not standardized owing to the need for various stent diameters and lengths. In addition, owners were an integral part of the decision-making process for whether a UMS or CMS was used owing to the potential financial burden should tissue ingrowth occur with a UMS and replacement with a CMS was needed. Owing to this lack of standardization, we were unable to develop an algorithm for what type of stent should be used in which circumstance. This is an area for further research.
The rate of success, defined as patency of the nasopharynx as determined by physical examination and clinical signs after stent placement, was high for both UMSs (20/30 [67%]) and CMSs (11/11), when compared with a single balloon dilatation procedure alone (8/27 [30%]). Notably, for most patients in which a stent was placed, balloon dilatation alone had previously failed. There were no immediate complications after UMS or CMS placement. Stent placement, therefore, appeared to be a safe and effective means for the immediate relief of nasopharyngeal obstruction secondary to NPS or INP.
In the present study, treatment with mitomycin C or triamcinolone after stent placement did not significantly reduce the incidence of tissue ingrowth, but cases that received these treatments were not randomized. Most commonly, the authors used these treatments when a covered stent was not being used or when a thick INP or similar lesion with a high risk of recurrence was being treated. A prospective randomized evaluation of these treatments would need to be performed to determine their usefulness.
Overall, 23 of the 34 (68%) patients in which stents were placed developed a complication. The complications that were observed were similar to those previously reported in other small case reports.6,8 Occurrence of complications was not significantly associated with whether a UMS or CMS was used. However, the types of complications differed between stent types. For the 30 patients in which UMSs were placed, tissue ingrowth (n = 10), chronic infection (7), and stent fracture (5) were the most common complications, whereas the complications most commonly seen in the 11 patients in which CMSs were placed were chronic infection (8) and development of an oronasal fistula (3). This difference was likely due to the open interstices of the UMSs, which allowed for tissue ingrowth. However, 1 benefit of UMSs is that the metallic mesh incorporates into the mucosa of the nasopharynx, avoiding a foreign body reaction and accumulation of mucus and subsequent infection. Covered metallic stents are coated with various types of materials such as silicone or polytetrafluoroethylene. Although the coating seems to prevent tissue ingrowth and stent fracture and bending, it does not allow mucosal incorporation of the stent edges, so that foreign material remains exposed within the nasopharynx.
Oronasal fistulas formed in 4 of the 30 patients in which a UMS was placed and in 3 of the 11 patients in which a CMS was placed. The cause of fistula formation was not clear but was likely associated with the stent edges moving along the surface of the palate with swallowing or chewing, ultimately resulting in breakdown of the tissue. Because this was most commonly seen in patients with lesions in the middle third of the nasopharynx, it is possible that this is a high-motion area of the soft palate. Given the variety of CMSs used, it was not possible to determine whether newer silicone coatings were less likely to be associated with oronasal fistula formation. Currently, the authors are further evaluating a novel covered SEMS with less outward radial force that is thought to cause less pressure on the palate and ultimately less tissue necrosis and breakdown.
It was surprising that only 1 of the CMSs migrated, as we had expected that because of decreased tissue incorporation of the stent, CMSs would be more likely to migrate. The small number of CMSs that were placed and the fact that most (7/11) CMSs were placed after failure of a UMS placement may account for this.
Development of complications was also not significantly associated with stent expansion type (BEMS vs SEMS); however, 20 of 27 of patients with a BEMS had a complication, compared with 3 of 7 patients with a SEMS. Over the past 3 years, the type of CMS being used for NPS or INP has changed to a self-expandable, silicone-covered stent, rather than a balloon-expandable, polytetrafluoroethylene-coated metallic stent. This seems to be resulting in fewer complications, particularly those associated with chronic infections and oronasal fistula. The thought is that the rigidity of a BEMS can cause more trauma to the palate, compared with an SEMS, which is more pliable and softer, with some recoil during palate motion. The rigidity of BEMSs may also make them more likely to migrate.
Development of complications was also not significantly associated with species, although 16 of 20 cats had complications with stents versus 7 of 14 dogs. This difference may have been related to the intrinsic nature of the disease, the underlying etiology of NPS or INP, or the type of stents used. The most common presumptive cause of NPS and INP in the cats in our study was chronic rhinitis or nasopharyngitis, but this was the underlying cause in only 1 dog. The most common presumptive cause of NPS or INP in the dogs was postanesthetic aspiration rhinitis. In these patients, the underlying chronic inflammatory or infectious disease process may have led to a higher likelihood of complications, chronic discharge, and chronic infections.
Lesion type (NPS vs INP) was not significantly associated with whether any complications occurred or with whether any specific complications occurred. Interestingly, however, overall outcome was significantly associated with lesion type, with outcome more likely to be successful for patients with NPS than INP. Subjectively, INP seemed to be more difficult to treat, which may have increased the likelihood of recurrence and owner dissatisfaction. As a result, we currently encourage considering using a covered SEMS in patients with INP, especially dogs, because all dogs with INP in which a CMS was placed had a successful outcome.
Although development of complications was not significantly associated with lesion location, patients with a lesion located in the middle third of the nasopharynx were more likely to have chronic infection, stent migration, and tissue ingrowth. This was suspected to be secondary to the anatomy of the nasopharynx and the motion of this region of the soft palate, compared with motion of the rostral third of the nasopharynx at the junction between the hard and soft palates. All patients with stent bending and exaggerated swallowing had a caudally located lesion, and these complications may have been a result of oropharyngeal irritation. More caudal placement of the stent allows for more ventral displacement of the soft palate into the oral cavity as a result of the rigidity and weight of the stent, causing exaggerated swallowing and stent bending.
Stent removal was necessary in 6 of 34 (18%) patients in the present study because of exaggerated swallowing, oronasal fistula development, chronic infection, or stent migration. Two of these patients went on to have a successful outcome following stent removal, which lends support to current work being done evaluating retrievable or temporary stents in cats. Stent removal was not significantly associated with species, lesion type, lesion location, or stent type in the present study. Thus, which patients are more likely to require stent removal remains to be determined.
Overall, a successful outcome was observed in 2 of the 4 patients with a CMS alone, 19 of the 23 patients with a UMS alone, and 4 of the 7 patients with both a UMS and a CMS. Stent type and ultimate procedure performed were not significantly associated with outcome. Despite the higher complication rate in cats, the success rate was significantly higher for cats (27/31 [87%]) than for dogs (9/15 [60%]). Outcome was also significantly associated with lesion location; however, regression analysis indicated this was biased by the fact that cats more commonly had a caudal lesion location and a successful outcome. This suggested the disease process may be less aggressive in cats than in dogs, which fits with the authors’ clinical impression. The reason cats developed more complications, despite being more likely to have a successful outcome, may be related to lesion anatomy or to species differences in implant tolerance and the propensity for upper respiratory tract infections. A larger sample size is needed to further elucidate the differences between cats and dogs observed in the present study.
The present study had many limitations, most of which were associated with its retrospective nature and the lack of standardization of which patients underwent balloon dilatation or stent placement, the particular type of stent used, and the type of coating for CMSs. There may have been a selection bias, in that the authors and clients may have been more inclined to place stents rather than perform repeated balloon dilatation procedures owing to the financial expense of multiple procedures and the fact that balloon dilatation had already failed in many patients and that clients were referred specifically for stent placement. All complications were included in our analysis, regardless of whether they could be corrected with medical management or required additional procedures to be performed. This may have resulted in a focus on complications, rather than outcome, following stent placement, as there was a high overall success rate. Many complications were considered relatively minor by the patients’ owners.
In conclusion, results suggested that NPS and INP in dogs and cats can be successfully treated with balloon dilatation or stent placement, but that there is a high risk of failure with balloon dilatation alone and a high risk of complications regardless of procedure. The likelihood of a successful outcome was higher with placement of a UMS or CMS; however, complications were also more likely. On the basis of our data, novel stents are being evaluated to determine whether they can maintain patency while avoiding some of the reported complications.
Acknowledgments
No third-party funding or support was received in connection with this study or the writing or publication of the manuscript. The authors declare that there were no conflicts of interest.
Presented in abstract form at the American College of Veterinary Internal Medicine Forum, Nashville, Tenn, June 2014.
ABBREVIATIONS
| BEMS | Balloon-expandable metallic stent |
| CMS | Covered metallic stent |
| INP | Imperforate nasopharynx |
| NPS | Nasopharyngeal stenosis |
| SEMS | Self-expandable metallic stent |
| UMS | Uncovered metallic stent |
Footnotes
6-mm gastroscope, Pentax, Westminster, Conn.
Flexible video bronchoscope, Olympus America, Center Valley, Pa.
Weasel Wire 0.025- or 0.035-inch hydrophilic angle-tipped guidewire, Infiniti Medical LLC, Menlo Park, Calif.
5F marker catheter, Infiniti Medical LLC, Menlo Park, Calif
Omnipaque-iohexol solution 240, GE Healthcare, Wauwatosa, Wis.
Vascular sheath, Infiniti Medical LLC, Menlo Park, Calif.
Renal Access Set, Infiniti Medical LLC, Menlo Park, Calif.
Vet Balloons, Infiniti Medical LLC, Menlo Park, Calif.
Cook inflation device, Cook Medical, Bloomington, Ind.
Vet Stent—Nasopharyngeal, Infiniti Medical LLC, Menlo Park, Calif.
Vascular stent, Cordis Corp, Fremont, Calif.
Wallstent Endoprosthesis, Boston Scientific, Marlborough, Mass.
Fluency covered stent, Bard International, New Providence, NJ.
Atrium cast covered stent, Atrium Medical Corp, Hudson, NH.
SAS, version 9.4, SAS Institute, Cary, NC.
References
1. Billen F, Day MJ, Clercx C. Diagnosis of pharyngeal disorders in dogs: a retrospective study of 67 cases. J Small Anim Pract 2006;47:122–129.
2. Talavera Lopez J, Josefa Fernandez del Palacio MA, Cano FG, et al. Nasopharyngeal stenosis secondary to soft palate dysgenesis in a cat. Vet J 2009;181:200–204.
3. Kirberger RM, Steenkamp G, Spotswood TC, et al. Stenotic nasopharyngeal dysgenesis in the Dachshund: seven cases (2002–2004). J Am Anim Hosp Assoc 2006;42:290–297.
4. Coolman BR, Marretta SM, McKiernan BC, et al. Choanal atresia and secondary nasopharyngeal stenosis in a dog. J Am Anim Hosp Assoc 1998;34:497–501.
5. Mitten RW. Nasopharyngeal stenosis in four cats. J Small Anim Pract 1988;29:341–345.
6. Berent AC, Weisse C, Todd K, et al. Use of a balloon-expandable metallic stent for treatment of nasopharyngeal stenosis in dogs and cats: six cases (2005–2007). J Am Vet Med Assoc 2008;233:1432–1440.
7. Hunt GB, Perkins MC, Foster SF, et al. Nasopharyngeal disorder of dogs and cats: a review and retrospective study. Compend Contin Educ Pract Vet 2002;24:184–198.
8. Cook AK, Mankin KT, Saunders AB, et al. Palatal erosion and oronasal fistulation following covered nasopharyngeal stent placement in two dogs. Ir Vet J 2013;66:8.
9. Allen HS, Broussard J, Noone KE. Nasopharyngeal diseases in cats: a retrospective study of 53 cases (1991–1998). J Am Anim Hosp Assoc 1999;35:457–461.
10. Glaus TM, Tomsa K, Reusch CE. Balloon dilation for the treatment of chronic recurrent nasopharyngeal stenosis in a cat. J Small Anim Pract 2002;43:88–90.
11. DeSandre-Robinson DM, Madden SN, Walker JT. Nasopharyngeal stenosis with concurrent hiatal hernia and megaesophagus in an 8-year-old cat. J Feline Med Surg 2011;13:454–459.
12. Griffon DJ, Tasker S. Use of a mucosal advancement flap for the treatment of nasopharyngeal stenosis in a cat. J Small Anim Pract 2000;41:71–73.
13. Novo RE, Kramek B. Surgical repair of nasopharyngeal stenosis in a cat using a stent. J Am Anim Hosp Assoc 1999;35:251–256.
14. Glaus TM, Gerber B, Tomsa K, et al. Reproducible and long-lasting success of balloon dilation of nasopharyngeal stenosis in cats. Vet Rec 2005;157:257–259.
15. Boswood A, Lamb CR, Brockman DJ, et al. Balloon dilatation of nasopharyngeal stenosis in a cat. Vet Radiol Ultrasound 2003;44:53–55.
16. Sidell D, Chhetri DK. CO2 laser ablation and balloon dilation for acquired nasopharyngeal stenosis: a novel technique. Laryngoscope 2011;121:1486–1489.
17. Berent AC, Kinns J, Weisse C. Balloon dilatation of nasopharyngeal stenosis in a dog. J Am Vet Med Assoc 2006;229:385–388.
18. De Lorenzi D, Bertoncello D, Comastri S, et al. Treatment of acquired nasopharyngeal stenosis using a removable silicone stent. J Feline Med Surg 2015;17:117–124.
