844 ScientificReports JAVMA,Vol235,No.7,October1,2009

SM

ALL

AN

IMA

LS

Esophageal bougienage refers to dilation of the esoph-agus with mechanical dilators, also known as rigid

dilators, push dilators, or bougies,1,2 and has been com-monly used for the treatment of BES in animals and peo-ple. The 2 main types of mechanical esophageal dilators currently available include freely passed mercury-filled bougies (eg, Maloney and Hurst dilators) and wire-guid-ed polyvinyl (eg, Savary-Gilliard dilators) or metal olive (eg, Eder-Puestow) bougies.2,3

In 1981, the use of balloon catheters for esophageal dilation in people was described,4 and balloon dilation rapidly gained popularity with gastroenterologists, as it was thought to be safer and more efficacious than bougie-nage.5,6 Balloon dilators allow radial forces to be applied to an area of esophageal stenosis, whereas mechanical dilators result in progressive proximal-to-distal applica-tion of longitudinal and radial forces. Despite this differ-ence, most randomized studies2,3 comparing the use of mechanical and balloon dilators in people have failed to show that balloon dilation was superior. Currently, both mechanical and balloon dilators are commonly used and recommended for the treatment of BES in people.2,7

In 1987, Burk et al8 described excellent outcomes following 1 to 2 episodes of balloon dilation in 6 dogs

Risk factors and outcome of bougienage for treatment of benign esophageal strictures

in dogs and cats: 28 cases (1995–2004)

Sally A. Bissett, bvsc, mvsc, dacvim; Julianne Davis, dvm; Katie Subler dvm; Laurel A. Degernes, dvm, mph, dabvp

Objective—Todetermineoutcomeofbougienagefortreatmentofbenignesophagealstric-ture(BES)indogsandcatsandidentifyriskfactorsforthecondition.Design—Retrospectivecaseseries.Animals—20dogsand8catswithBES.Procedures—Medical recordswere reviewedfor informationonsignalment,clinical fea-tures,andoutcome.Long-termoutcomeinformationwasobtainedwithaquestionnaire.Results—Esophagealbougienagewasperformedwithdilatorsrangingfrom5to15mmindiameter;mediannumbersofbougienageprocedureswere3(dogs)and4.5(cats).Agoodoutcome,definedastoleranceofsolidfoodwithregurgitationlessthanonceaweek,wasachieved in 14dogs and6 cats.Complicationswereminimal,withnonfatal esophagealperforationoccurring in1cat.Fourdogsand1catwereeuthanizedordiedbecauseofesophagealdisease.DogswithBESweremorelikelytobefemale,havearecenthistoryofgeneralanesthesia,havereceivedanantimicrobialorally,orhaveahistoryofvomitingthanweredogsinareferencepopulation.CatswithBESweremorelikelytohavearecenthistoryofgeneralanesthesia,vomiting,orgastrointestinal tract trichobezoars thanwerecatsinareferencepopulation.Doxycycline-inducedesophagitiswasthesuspectedcauseofBESin3cats.AlthoughgeneralanesthesiawasassociatedwithdevelopmentofBESin18dogsand5cats,concurrentpotentialcausesofesophagealinjurywerecommon.Conclusions and Clinical Relevance—Results suggested that esophageal bougienagewasasafeandeffectivetreatmentformostdogsandcatswithBES,withoutcomessimilartothosereportedforballoondilation.(J Am Vet Med Assoc2009;235:844–850)

and cats with BES and concluded that balloon dilation was a superior method of treatment. Since that time, most reports9–17 describing esophageal dilation for treatment of BES in dogs and cats have been limited to use of balloon dilation, although a few reports10,18,19 have described using an endoscope tip as a mechanical dilator. Other reports20–23 refer to the use of esophageal bougienage in dogs with BES, but do not describe the dilators used, the procedure itself, or the outcome in detail. Despite a substantial cost advantage of mechani-cal dilation, compared with balloon dilation,7 esopha-geal bougienage appears to be uncommon in dogs and cats at the present time.

Gastroesophageal reflux associated with general anesthesia is reportedly the most common cause of BES in dogs and cats.10,12,14,15 However, any substantial chemical or mechanical injury to the esophageal mu-cosa has the potential to result in stricture formation. Esophageal strictures can be devastating for pet own-ers, as they can be costly to treat and associated with a poor prognosis. In previous studies,10,12,14,20 10% to 30% of affected dogs and cats died or were euthanized and 15% to 30% could only be fed liquids long term, despite multiple episodes of esophageal dilation.

From the Department of Clinical Sciences, College of Veterinary Medi-cine, North Carolina State University, Raleigh, NC 27606.

Address correspondence to Dr. Bissett (sally_bissett@ncsu.edu).

Abbreviation

BES Benignesophagealstricture

JAVMA,Vol235,No.7,October1,2009 ScientificReports 845

SM

ALL A

NIM

ALS

The primary purpose of the study reported here was to determine outcome of bougienage for treatment of BES in dogs and cats. Secondarily, we wanted to iden-tify risk factors associated with development of BES.

Materials and Methods

Case selection criteria—The medical records and radiology database of the North Carolina State Uni-versity Veterinary Teaching Hospital were searched to identify dogs and cats examined between January 1995 and December 2004 in which a diagnosis of BES had been made. Cases were included in the study if the di-agnosis of BES had been confirmed by means of esopha-goscopy and the stricture had been treated by means of bougienage.

For identification of risk factors associated with de-velopment of BES, a reference population was obtained by randomly selecting, for each dog or cat with BES, 3 dogs or 3 cats admitted to the hospital on the same day as the dog or cat with BES. Reference dogs and cats were selected by the use of computer-generated random numbers from a numbered list of all dogs or cats admit-ted to the small animal hospital on dates corresponding to the initial examination of each dog or cat with BES. Dogs and cats with esophageal disease were excluded from the reference population.

Medical records review—Data retrieved from the medical records of dogs and cats with BES included date of admission, signalment, clinical signs at the time of admission, any history of events that occurred prior to the onset of clinical signs and could have been as-sociated with esophageal injury (eg, general anesthesia, oral administration of drugs, vomiting, and gastrointes-tinal tract foreign material), physical examination find-ings, results of esophageal imaging (ie, radiography, fluoroscopy, and endoscopy), dates and details of all bougienage procedures, complications associated with esophageal bougienage, and outcome. Esophagoscopy images were reviewed by a single board-certified inter-nist (SAB) for consistency in interpretation of mucosal abnormalities.

Data retrieved from the medical records of reference dogs and cats included in the study consisted of age, sex, breed, body weight, and any history of events that could have been associated with esophageal injury.

Esophageal bougienage procedure—For all dogs and cats with BES, esophageal bougienage was per-formed with Savary-Gilliard bougies.a Following en-doscopic inspection of the stricture site, diameter of the stricture was estimated by comparison with vari-ous-sized bougies, endoscopic forceps, or the tip of a 9.8-mm gastroscope. A guide wire was then advanced down the esophagus past the stricture site, and a well-lubricated bougie was passed over the guide wire until its maximum diameter had passed the stricture site. This bougie was then removed, and a slightly larger bougie was passed over the guide wire and through the stric-ture site. In general, the initial bougie was selected to be the same size as or up to 2 mm larger than the estimated diameter of the stricture. Subsequent bougies were 1 to 3 mm larger than the previous bougie, and typically,

2 to 4 bougies, depending on the degree of mucosal trauma and friction encountered, were passed during each esophageal bougienage session. To avoid excessive mucosal injury, the largest bougie selected was usually ≤ 6 mm larger than the initial estimated diameter of the stricture. Bougies were typically passed blindly over the guide wire, but when possible, passage was monitored endoscopically. The esophageal mucosa was inspected following passage of each bougie, and the session was discontinued if excessive mucosal bleeding or tearing was seen. Additional esophageal bougienage sessions were recommended by the primary clinician in charge of each case as needed; additional bougienage sessions were usually recommended if the estimated diameter of the stricture site remained < 10 mm or if clinical signs persisted or recurred.

Outcome—Information on long-term outcome was obtained through a telephone questionnaire ad-ministered to owners. Owners were asked to provide information on frequency of regurgitation, consistency of food tolerated, cause of death (if known), whether they considered treatment to have been successful, and how satisfied they were with the outcome. When avail-able, medical records were also obtained from referring veterinarians.

On the basis of information obtained from medi-cal records and telephone questionnaires, dogs and cats with BES were categorized according to potential cause of BES (ie, general anesthesia, oral antimicrobial administration, vomiting, or trichobezoars [cats only]) and outcome following bougienage. Outcome was clas-sified as good if the animal was able to tolerate solid food with regurgitation less than once a week and was classified as poor otherwise (eg, animal regurgitated more than once a week if fed solid or liquid food or was able to tolerate liquid food with regurgitation less than once a week). For purposes of the present study, solid food was defined as kibble, kibble soaked in water, and canned food, and liquid food was defined as any food with a consistency of gruel.

Statistical analysis—Because data for most vari-ables were not normally distributed, the median and 5th and 95th percentiles are reported. Logistic regres-sion was used to identify risk factors potentially as-sociated with BES; odds ratios and 95% confidence intervals were calculated for factors significantly (P ≤ 0.05) associated with BES. Age was categorized as young (< 6 years old for dogs and < 8 years old for cats) or old (≥ 6 years old for dogs and ≥ 8 years old for cats). For dogs, body weight was categorized as small-medium (< 26 kg [58.5 lb]) or large (≥ 26 kg). The χ2 test was used to compare distributions for age, sex, body weight (dogs only), and a history of general anesthesia, vomiting, antimicrobial use (all types com-bined and β-lactams or enrofloxacin [dogs only]), or gastrointestinal tract trichobezoars (cats only) between animals with BES and animals in the reference popu-lation. Because of potential confounding of possible risk factors for BES (eg, a history of general anesthesia, vomiting, and antimicrobial administration), multivari-ate logistic regression was used to identify risk factors for BES in dogs. However, the small number of cats with

846 ScientificReports JAVMA,Vol235,No.7,October1,2009

SM

ALL

AN

IMA

LS

BES included in the study precluded development of a multivariate model. Finally, for the dogs, the χ2 test was used to determine whether possible risk factors (eg, a history of general anesthesia, vomiting, or antimicrobi-al administration) or physical property of the stricture (eg, location, initial stricture diameter, and presence of concurrent esophagitis) were associated with outcome. All analyses were performed with standard software.b Values of P ≤ 0.05 were considered significant.

Results

Twenty dogs and 8 cats with BES met the criteria for inclusion in the study. The reference population consisted of 60 dogs and 24 cats without any evidence of esophageal disease.

Signalment—Of the 20 dogs with BES, 3 were of mixed breeding, with the remaining 17 representing 13 breeds (Boston Terrier, Chow Chow, Labrador Retriev-er, and Toy Poodle [2 dogs each] and Bulldog, Cocker Spaniel, Golden Retriever, Great Dane, Jack Russell Terrier, Old English Sheepdog, Pomeranian, Shetland Sheepdog, and Shih Tzu [1 dog each]). The 8 cats with BES included 3 domestic shorthairs, 3 domestic long-hairs, 1 Manx, and 1 Tonkinese. Notably, fewer cats

with BES were domestic shorthairs (3/8 [38%]) and more cats were domestic longhairs (3/8 [38%]) than cats in the reference population (19/24 [79%] and 1/24 [4.2%], respectively). Other substantial differences in breed distribution between animals with BES and ani-mals in the reference populations were not identified. All 3 of the domestic longhair cats with BES had a his-tory of gastrointestinal tract trichobezoars.

Age and body weight did not differ significantly between animals with BES and animals in the reference populations (Table 1). However, dogs with BES were significantly more likely to be female than were dogs in the reference population.

Risk factors for BES—Results of multivariate logis-tic regression indicated that dogs with BES were more likely to be female, have a recent history of general an-esthesia, have received an antimicrobial orally, or have a history of vomiting than were dogs in the reference population (Table 2). Univariate analyses indicated that cats with BES were more likely to have a recent history of general anesthesia, vomiting, or gastrointes-tinal tract trichobezoars than were cats in the reference population (Table 3).

Dogs Cats

Affected Reference Affected ReferenceVariable (n=20) (n=60) (n=8) (n=24)

Age (y) 6.8 (1.0–11.5) 8.0 (0.5–16) 4.5 (0.3–15.5) 8.0 (2.5–17)Sex Male 5 (25) 39 (65) 5 (63) 14 (58) Female 15 (75) 21 (35) 3 (37) 10 (42)Weight 17.2 (2.0–48.0) 22.7 (3.2–65.5) 3.8 (1.7–4.8) 4.4 (3.2–7.0) (kg)

Data are given as median (5th percentile–95th percentile) or as number (%) of animals. Sex distribution differed significantly (P 0.05) between affected dogs and dogs in the reference population.

Table1Age,sexdistribution,andbodyweightofdogsandcatswithBES(affected)andofdogsandcatsinareferencepopula-tionwithoutevidenceofesophagealdisease.

Univariateanalysis Multivariateanalysis

Variable OR 95%CI Pvalue OR 95%CI Pvalue

Age (y) 1.2 0.4–3.3 0.79 NA NA NASex 5.6 1.8–17.5 0.01 46.5 2.4–903.5 0.01Weight (kg) 1.3 0.4–3.8 0.68 NA NA NAHistory of: Recent general anesthesia 83.3 14.9–500 0.01 333.3 15.2–1,000 0.01 Oral antimicrobial administration 10.4 3.3–33.3 0.01 13.0 1.1–166.7 0.04 Oral administration of a β-lactam 13.2 3.7–47.6 0.01 NA NA NA Oral administration of enrofloxacin 5.5 1.6–21.3 0.01 NA NA NA Vomiting 6.0 1.4–24.4 0.01 27.0 1.1–1,000 0.05

Analyses were performed with data for 20 dogs with BES and 60 dogs in a reference population without esophageal disease. The odds ratio represents the odds of BES among dogs 6 years old, compared with the odds among dogs 6 years old; the odds of BES among female dogs, compared with the odds among male dogs; the odds among dogs weighing 26 kg, compared with the odds among dogs weighing 26 kg; or the odds among dogs with a history of the factor of interest, compared with the odds among dogs without any such history.

CI = Confidence interval. NA = Not applicable (factor was not included in multivariate model). OR = Odds ratio.

Table2Resultsofunivariateandmultivariatelogisticregressionanalysisoffactorspotentiallyassoci-atedwithdevelopmentofBESindogs.

Variable OR 95%CI Pvalue

Age (y) 3.6 0.6–22.2 0.17Sex 0.8 0.2–4.4 0.84History of: Recent general anesthesia 18.2 2.4–149.2 0.01 Oral antimicrobial administration 0.4 0.1–1.9 0.23 Vomiting 6.3 1.1–35.7 0.04 Gastrointestinal tract trichobezoar 11.0 1.5–83.3 0.02

Analyses were performed with data for 8 cats with BES and 24 cats in a reference population without esophageal disease. The odds ratio represents the odds of BES among cats 8 years old, compared with the odds among cats 8 years old; the odds of BES among female cats, compared with the odds among male cats; or the odds among cats with a history of the factor of interest, com-pared with the odds among cats without any such history.

See Table 2 for remainder of key.

Table3Resultsofunivariatelogisticregressionanalysisoffac-torspotentiallyassociatedwithdevelopmentofBESincats.

JAVMA,Vol235,No.7,October1,2009 ScientificReports 847

SM

ALL A

NIM

ALS

Clinical signs of BES—Median duration of clinical signs of esophageal disease prior to referral was 24 days (5th percentile, 5 days; 95th percentile, 47 days) for dogs and 19 days (5th percentile, 3 days; 95th percentile, 36 days) for cats. Reported clinical signs and physical exam-ination findings for animals with BES included regurgita-tion (20 dogs and 8 cats), gagging associated with eating (11 dogs), ptyalism (9 dogs and 1 cat), exaggerated swal-lowing and lip licking (8 dogs and 6 cats), odynophagia (3 dogs and 2 cats), weight loss (15 dogs), coughing (7 dogs), and decreased appetite (3 dogs and 5 cats).

Causes of esophageal injury—Eighteen (90%) dogs and 5 (63%) cats with BES had undergone general anesthe-sia for various reasons between 1 and 17 days prior to the onset of clinical signs of esophageal disease. Procedures performed in animals that had undergone general anes-thesia included abdominal surgery (8 dogs [including 3 that underwent ovariohysterectomy] and 2 cats), removal of pharyngeal or esophageal trichobezoars (4 cats, includ-ing 1 cat that underwent abdominal surgery), skin mass removal (4 dogs), aural surgery (3 dogs), and thoracic sur-gery, orthopedic surgery, and contrast myelography (1 dog each). Thirteen of the dogs and all 5 cats with a history of having recently undergone general anesthesia had a his-tory of other events that might have caused esophageal injury. Only 6 of the 60 (10%) dogs and 2 of the 24 (8%) cats in the reference populations had a history of undergo-ing general anesthesia within 30 days prior to admission.

Two dogs and 3 cats with BES did not have any his-tory of having recently undergone general anesthesia. However, all 5 had a history of other events that might have caused esophageal injury, including vomiting and oral enrofloxacin administration in the 2 dogs and oral doxycycline administration in the 3 cats.

Fifteen (75%) dogs and 5 (63%) cats with BES had a history of oral drug administration within 1 to 17 days prior to the onset of clinical signs of esophageal dis-ease. Antimicrobials were administered in 14 dogs and 5 cats, including β-lactam antimicrobials (ie, cephalexin, amoxicillin, and amoxicillin-clavulanic acid) in 11 dogs and 3 cats, enrofloxacin in 7 dogs, doxycycline in 3 cats, clindamycin in 1 dog, and metronidazole in 1 dog. Six dogs and 1 cat received 2 or 3 antimicrobials at the same time. All 15 dogs and 3 of the cats with a history of oral drug administration also had a history of other events that might have caused esophageal injury. Eleven of the 60 (18%) dogs and 9 of the 24 (38%) cats in the refer-ence populations had a history of oral antimicrobial ad-ministration within 30 days prior to admission, although none of these cats had received doxycycline.

Six (30%) dogs and 5 (63%) cats with BES had a history of vomiting prior to the onset of clinical signs of esophageal disease. Vomiting was classified as acute (1 to 12 days prior to the onset of regurgitation) in 5 dogs and 2 cats and was classified as chronic (> 3 weeks prior to the onset of regurgitation) in 1 dog and 3 cats. Small pieces of plastic or bone were identified in the vomitus from 3 dogs, whereas trichobezoars were iden-tified in the vomitus from 4 (50%) cats. Three cats with a history of vomiting had had esophageal trichobezoars removed prior to referral for evaluation of regurgitation. Two of these cats had endoscopic evidence of esophagi-tis at the time of trichobezoar removal. The fourth cat

had had a trichobezoar removed from its pharynx prior to referral for evaluation of regurgitation. All 6 dogs and 5 cats with a history of vomiting also had a history of other events that might have caused esophageal injury. Four of the 60 (7%) dogs and 5 of the 24 (21%) cats in the reference populations had a history of vomiting, and 2 (8%) cats had a history of vomiting trichobezoars within 30 days prior to admission.

The only other cause of esophageal stricture that was identified was esophageal surgery in 1 dog. This dog had had a BES prior to esophageal surgery, and multiple episodes of balloon dilation of the stricture at another referral hospital had failed to yield any improvement.

Diagnostic imaging—Thoracic radiographs ob-tained from 16 dogs and 6 cats with BES were available for review. Esophageal abnormalities were observed in 9 dogs and 4 cats and included focal esophageal gas (7 dogs and 4 cats), fluid in the caudal portion of the esophagus (2 dogs), and megaesophagus (1 dog). None of the animals had radiographic evidence of aspiration pneumonia at the time of admission.

Videofluoroscopic esophagraphy was performed in 14 dogs and 5 cats with liquid barium and barium mixed with solid food (10 dogs and 2 cats), liquid barium only (4 dogs), or iohexol (3 cats). Esophageal abnormalities were identified in 13 dogs and all 5 cats and included diffuse (5 dogs) or segmental (6 dogs and 5 cats) esophageal dys-motility, segmental esophageal dilation (7 dogs and 4 cats), and focal narrowing of contrast material suggestive of an esophageal stricture (6 dogs and 4 cats). In 1 dog, results of esophagraphy were unremarkable despite administration of liquid barium and barium mixed with solid food.

Esophagoscopy was performed in all 28 dogs and cats. A single stricture was observed in 10 dogs and 8 cats, and 2 strictures were observed in the remaining 10 dogs. The strictures were located cranial to the thoracic inlet in 2 dogs (3 strictures) and 1 cat, between the thoracic inlet and the heart base in 7 dogs (10 strictures) and 6 cats, and caudal to the heart base in 13 dogs (17 stric-tures) and 1 cat. Stricture diameter was estimated for the smallest or most cranial stricture when 2 strictures were present. Stricture diameter was estimated to be ≤ 5 mm in 9 dogs and 5 cats, > 5 but ≤ 9.8 mm in 7 dogs and 3 cats, and > 9.8 mm in 3 dogs (estimated stricture diameter was not reported in 1 dog). Although most strictures appeared to be < 1 cm long on endoscopic images, stricture length was not reliably reported. Irregular, hyperemic, or erosive esophageal mucosa consistent with esophagitis was de-tected in 10 dogs and 3 cats.

Esophageal bougienage—All 28 dogs and cats un-derwent bougienage at least once for treatment of the esophageal stricture, but a total of 125 esophageal bou-gienage sessions were performed. Median numbers of bougienage episodes were 3 (5th percentile, 1.9; 95th percentile, 10.0) for the dogs and 4.5 (5th percentile, 1.0; 95th percentile, 9.0) for the cats. Only 1 dog and 2 cats had a single esophageal bougienage session. For animals in which > 1 bougienage session was performed, treat-ment occurred over a median of 17 days (5th percentile, 5 days; 95th percentile, 96 days) in the dogs and 21 days (5th percentile, 13 days; 95th percentile, 77 days) in the cats, with a range of 2 to 48 days between sessions. Di-

848 ScientificReports JAVMA,Vol235,No.7,October1,2009

SM

ALL

AN

IMA

LS

ameter of bougies used during the initial bougienage ses-sion ranged from 5 to 12 mm for most dogs and from 5 to 9 mm for most cats. Diameter of bougies used for the final bougienage session ranged from 9 to 15 mm for the dogs and from 9 to 12 mm for the cats. The largest bou-gie used was 15 mm in diameter.

Complications associated with esophageal bougie-nage were minimal. Two dogs developed mild aspiration pneumonia associated with regurgitation at the time of anesthetic induction. In 1 cat, pneumomediastinum de-veloped following esophageal bougienage but a gastric trichobezoar was also removed endoscopically during this session. The cat was assumed to have an esophageal perforation. Treatment consisted of percutaneous endo-scopically assisted placement of a gastric feeding tube and antimicrobial administration; the cat survived and underwent additional esophageal bougienage sessions without problems. Mucosal trauma following esopha-geal bougienage sessions was subjectively evaluated to be mild (no to small amount of hemorrhage) in 11 dogs and 6 cats, moderate (hemorrhage with small mucosal tears) in 2 dogs, and severe (hemorrhage with large mu-cosal tears) in 1 dog. Missing or poor-quality endoscopic images precluded evaluation of the severity of mucosal trauma following bougienage in 6 dogs and 2 cats.

Dogs and cats were treated with a variety of medica-tions following esophageal bougienage, including orally administered glucocorticoids (18 dogs and 8 cats), proton pump inhibitors or H

2 receptor antagonists (18 dogs and

5 cats), sucralfate (14 dogs and 4 cats), metoclopramide or cisapride (16 dogs and 4 cats), and various antimicrobi-als (8 dogs and 4 cats). Although standard drug dosages were used, the duration of administration was variable, ranging from 7 days to > 6 weeks. None of the dogs and cats received any glucocorticoid injections at the site of the stricture. Six dogs and 4 cats had a percutaneous gastric feeding tube placed at the time of esophageal bougienage.

Outcome—Owners of all 28 dogs and cats were suc-cessfully contacted and completed the telephone question-naire regarding long-term outcome. Median time from the last esophageal bougienage session to completion of the questionnaire was 7 years (5th percentile, 1.6 years; 95th percentile, 10.8 years). Ten (50%) dogs and 4 (50%) cats were alive at the time the questionnaire was completed; follow-up time for these animals ranged from 1.3 to 9.6 years. Four (20%) dogs and 1 (13%) cat had been eutha-nized or died as a direct consequence of the esophageal disease between 3 weeks and 6 months after the last bou-gienage session. Continued regurgitation following 2 or 3 esophageal bougienage sessions was the reason for eutha-nasia of 3 of the dogs. The cat was euthanized because of continued regurgitation following 10 esophageal bougie-nage sessions. The fourth dog had a good outcome for 6 months following 5 esophageal bougienage sessions but died suddenly following an episode of suspected esopha-geal obstruction or injury characterized by an acute onset of gagging, ptyalism, and regurgitation in association with eating that led to severe respiratory distress. The remain-ing 6 dogs and 3 cats were euthanized for unrelated ill-nesses or were reported to have died of old age > 6 months after the last esophageal bougienage session.

Outcome of esophageal bougienage was classified as good in 14 (70%) dogs and 6 (75%) cats and poor in 6

(30%) dogs and 2 (25%) cats. Four of the dogs and 2 of the cats with a good outcome ate dry food (kibble), and the remaining 10 dogs and 4 cats ate canned food with minimal regurgitation (ie, regurgitation less than once a week). Seven dogs and 4 cats with minimal regurgitation when fed canned food were never fed dry food following diagnosis of a BES. Two of the dogs and 1 of the cats with a poor outcome were fed liquid food with minimal regur-gitation but did not tolerate solid food. Two of the other dogs with a poor outcome regurgitated more often than once a week when fed solid food (canned food or kibble soaked in water) but were never fed liquid food. The re-maining 2 dogs and 1 cat with a poor outcome regurgitat-ed more than once a week despite being fed a liquid diet. Suspected cause of the esophageal stricture (ie, general anesthesia, antimicrobial use of any kind, or vomiting), stricture location (cranial vs caudal half of the esophagus), initial stricture diameter (≤ 5 vs > 5 mm), and presence of concurrent esophagitis did not differ significantly between dogs with a good versus a poor outcome.

Owners of 16 (57%) animals thought their pets were greatly improved following esophageal bougienage, own-ers of 9 (32%) animals thought their pets were mildly or moderately improved, and owners of 3 (11%) animals thought their pets were not improved at all. Owners of 19 (68%) animals indicated that they were overall satis-fied with the outcome of esophageal bougienage, where-as owners of 6 (21%) animals indicated that they were slightly satisfied, and owners of 3 (11%) animals indi-cated that they were not satisfied at all.

Discussion

Results of the present study suggested that esopha-geal bougienage was a safe and effective treatment for naturally occurring BES in dogs and cats. Outcome was classified as good (ie, animal tolerated solid food with re-gurgitation less than once a week) in 14 of the 20 (70%) dogs and 6 of the 8 (75%) cats. This was similar to per-centages of animals with good outcomes following bal-loon dilation in previous reports.12,14 In 1 study,12 16 of 22 (73%) animals tolerated solid food with only occasional regurgitation; in the other,14 14 of 19 (74%) animals tolerated solid food without regurgitation. Importantly, both of the previous studies12,14 involving balloon dila-tion excluded dogs and cats that were euthanized or died for various reasons, which may inflate the reported suc-cess rates. In contrast, in the present report, 3 dogs that were euthanized approximately 3 weeks following 2 to 3 esophageal bougienage sessions were classified as having a poor outcome and included in calculation of the suc-cess rate. If these 3 dogs had been excluded, the success rate would have been 82% (14/17) for the dogs. Although the outcome of esophageal bougienage in dogs and cats in the present study was similar to previously reported outcomes of balloon dilation, median numbers of bou-gienage sessions performed in the present report (3 for dogs and 4.5 for cats) were greater than median numbers of dilation sessions in those previous reports.12,14 Infor-mation on outcome for dogs and cats with BES published elsewhere8,10,15,20,21,23 could not be directly compared with results of the present report because there were insuf-ficient details regarding the consistency of food tolerated to make meaningful comparisons.

JAVMA,Vol235,No.7,October1,2009 ScientificReports 849

SM

ALL A

NIM

ALS

In the authors’ experience, it is common for veterinari-ans to believe that esophageal bougienage is associated with more complications (eg, esophageal perforation and diver-ticulum formation) than balloon dilation.8,24 However, this has not been found to be the case for people with BES,25,26 and esophageal perforation rate in the present study was low, both as a function of number of cases (1/28 [3.6%]) and as a function of number of bougienage sessions (1/125 [0.8%]). In addition, the esophageal perforation rate in the present study compared favorably with rates reported fol-lowing balloon dilation in dogs and cats (3.6% to 9% on a per-case basis and 2% to 3% on a per-session basis).12,14,15 High esophageal perforation rates, however, have been re-ported following the use of freely passed bougies (ie, bou-gies that are passed without placement of a guide wire) in people with complex strictures.27 Because most dogs and cats with BES are likely to have complex strictures (ie, strictures < 12 mm in diameter and asymmetric or tortu-ous strictures), the use of freely passed bougies should probably be avoided in dogs and cats. Factors other than dilation method (bougienage vs balloon dilation), such as cause of the stricture, severity of esophagitis, presence of a diverticulum, operator experience, size of the dilator rela-tive to size of the stricture and the animal, and amount of dilation pressure used, likely are important factors associ-ated with esophageal perforation in dogs and cats. A po-tential explanation for the low esophageal perforation rate in the present study is the modest maximum diameter of bougies that were used (15 mm), compared with the 18- to 20-mm-diameter balloons used in previous reports.12,14,15 Less serious complications such as moderate to severe mu-cosal trauma (tears and hemorrhage) and aspiration pneu-monia were identified in only 3 of 20 (15%) animals and 2 of 28 (7%) animals in the present study, respectively. This degree of mucosal trauma is similar to that reported follow-ing balloon dilation,12,14 and anesthetic technique, rather than the dilation procedure itself, appeared to be associated with the development of aspiration pneumonia.

An additional objective of the present study was to evaluate risk factors associated with development of BES in dogs and cats. Although signalment of and clini-cal abnormalities in dogs and cats with BES in the pres-ent study were largely consistent with findings reported previously,10,12,14,15,20 the comparison of affected animals with reference populations of unaffected animals allowed us to identify factors associated with development of BES. Specifically, female dogs in the present study were signifi-cantly more likely to have a BES than were male dogs. To our knowledge, only 3 other studies10,15,20 have described a predominance of females among dogs with BES, and this finding was attributed to the influences of progesterone and estrogen or to the high number of dogs that underwent ovariohysterectomy. However, few dogs had an ovariohys-terectomy in the present study, and the role of female sex hormones on lower esophageal sphincter function in dogs is unknown. Furthermore, gender does not appear to be a risk factor for gastroesophageal reflux in people.28 There-fore, an explanation for the female predisposition to BES in dogs is not readily apparent at this time.

A history of recent general anesthesia was also found to be a significant risk factor for development of BES in dogs and cats in the present study. This was not surprising given the high prevalence of gastroesophageal reflux in dogs un-

dergoing general anesthesia29,30 and numerous published re-ports10,12,14,15,23 linking anesthesia with BES. The prevalence of anesthesia-associated BES appears to be lower in cats than in dogs, however, and all of the cats with a history of recent anesthesia in the present study had other potential causes of BES. Because animals with a history of general anesthesia in the present study often also had a history of other potential causes of esophageal injury (eg, vomiting or antimicrobial administration), it was not possible to confirm the cause of stricture formation in many cases, and multiple factors may be important in the progression of esophageal injury to esophageal stricture formation.

Doxycycline-induced esophagitis was the suspected cause of BES in 3 of the 8 cats in the present study. How-ever, we were not able to determine whether it was a risk factor because of the small number of cats. We suspect a causal relationship between doxycycline and esophageal stricture formation because of the close temporal relation-ship between drug administration and the onset of clinical signs of esophageal disease and the failure to identify an-other potential cause of esophageal injury in 2 of the 3 cats that received doxycycline. Although > 70 drugs have been associated with drug-induced esophagitis in people,31 as far as the authors are aware, drug-induced esophagitis and BES have only been reported for cats receiving doxycycline or clindamycin orally.9,13,32 One dog in the present study had received clindamycin orally, but this dog was also being treated with enrofloxacin and had recently undergone gen-eral anesthesia. Interestingly, oral antimicrobial administra-tion was a significant risk factor for BES in dogs reported here. Although uncommon, ciprofloxacin and penicillin have been reported to cause esophagitis in people.31,33,34 All dogs in the present study with a history of oral anti-microbial administration prior to the onset of clinical signs of esophageal disease also had a history of other potential causes of esophageal injury. Therefore, further investigation is required to determine whether enrofloxacin or β-lactams are truly capable of inducing esophagitis and BES in dogs.

In previous studies10,12,14,15 of dogs and cats with BES, vomiting and esophageal foreign bodies were reported in up to 25% and 15% of cases, respectively. Distinguishing vomiting from regurgitation can be problematic in retro-spective studies of esophageal disease, but vomiting was only recorded in the present study if it occurred prior to the onset of regurgitation and was clearly defined in the medical record. Therefore, percentages of dogs (6/20 [30%]) and cats (5/8 [63%]) with a history of vomiting reported in the present study may have been underesti-mates of the true percentages. Importantly, however, all of the animals with a history of vomiting also had a history of other potential causes of esophageal injury. Although none of the dogs in the present study had a history of esophageal foreign bodies, 4 of the 8 cats had a history of pharyngeal or esophageal trichobezoars. It seems likely that physical abrasion and chemical injury secondary to vomiting played an important role in mucosal dam-age and stricture formation in these cases. Interestingly, esophageal trichobezoars have been described in 4 other cats with BES12,15,17 and may represent an underappreci-ated cause of esophagitis and BES in cats. In agreement with a report35 of intestinal trichobezoar obstructions, most of the cats with trichobezoars in the present study were domestic longhairs.

850 ScientificReports JAVMA,Vol235,No.7,October1,2009

SM

ALL

AN

IMA

LS

Benign esophageal strictures secondary to nonpeptic causes, strictures with a particularly narrow diameter, and strictures associated with ongoing symptoms of reflux have all been associated with worse outcomes in people.7,36 In agreement with findings of a previous study,12 however, we were unable to identify any relationship between outcome and potential cause of BES, stricture location, initial stric-ture size, or presence of esophagitis for dogs in the present study, although the low number of cases likely inhibited our ability to detect any such effect. We chose not to evaluate potential relationships between number of bougienage ses-sions and physical properties of the stricture or outcome because the number of bougienage sessions did not always reflect the clinical response and was dictated to some extent by owner preferences and financial limitations.

The technique for performing esophageal bougienage described in the present report is relatively simple, other than the endoscopic skills needed, and was partially based on our experience and guidelines for esophageal bougienage in peo-ple.2 Although there remain unanswered questions as to the optimal size and number of bougies to use during any partic-ular session, the procedure that we used appeared to be safe and effective for most of our dogs and cats. One advantage of esophageal bougienage, compared with balloon dilation, is that bougies are relatively cheap and durable, compared with balloon dilators. The Savary-Gilliard dilators used in the pres-ent study were approximately 20 years old, and the current price for a set of bougies ranging from 5 to 15 mm in diame-ter and a guide wire is approximately $2,000. By contrast, the current price for a disposable balloon dilator is approximately $145. Although disposable balloon dilators are meant to be single use, they can often be used several times. Regardless, both esophageal bougienage and balloon dilation appear to provide reasonable outcomes for dogs and cats with BES, and there is no evidence to support concerns that bougienage is associated with a higher risk of serious complications.

a. Savary-Gilliard Dilators, Cook Endoscopy, Winston Salem, NC.b. SAS, version 9.1.3, SAS Institute Inc, Cary, NC.

References1. Earlam R, Cunha-Melo JR. Benign oesophageal strictures: histori-

cal and technical aspects of dilatation. Br J Surg 1981;68:829–836.2. Spechler SJ. American Gastroenterological Association medi-

cal position statement on treatment of patients with dysphagia caused by benign disorders of the distal esophagus. Gastroenter-ology 1999;117:229–233.

3. Guelrud M. Management of benign esophageal strictures. UpToDate Web site. Available at: www.uptodate.com. Accessed Oct 10, 2008.

4. London RL, Trotman BW, DiMarino AJ Jr, et al. Dilatation of severe esophageal strictures by an inflatable balloon catheter. Gastroenterology 1981;80:173–175.

5. Lindor KD, Ott BJ, Hughes RW Jr. Balloon dilatation of upper digestive tract strictures. Gastroenterology 1985;89:545–548.

6. McBride MA, Ergun GA. The endoscopic management of esopha-geal strictures. Gastrointest Endosc Clin N Am 1994;4:595–621.

7. Riley SA, Attwood SE. Guidelines on the use of oesophageal dilatation in clinical practice. Gut 2004;53(suppl 1):i1–i6.

8. Burk RL, Zawie DA, Garvey MS. Balloon catheter dilation of in-tramural esophageal strictures in the dog and cat: a description of the procedure and a report of six cases. Semin Vet Med Surg (Small Anim) 1987;2:241–247.

9. German AJ, Cannon MJ, Dye C, et al. Oesophageal strictures in cats associated with doxycycline therapy. J Feline Med Surg 2005;7:33–41.

10. Adamama-Moraitou KK, Rallis TS, Prassinos NN, et al. Benign

esophageal stricture in the dog and cat: a retrospective study of 20 cases. Can J Vet Res 2002;66:55–59.

11. McGrotty YL, Knottenbelt CM. Oesophageal stricture in a cat due to oral administration of tetracyclines. J Small Anim Pract 2002;43:221–223.

12. Leib MS, Dinnel H, Ward DL, et al. Endoscopic balloon dilation of benign esophageal strictures in dogs and cats. J Vet Intern Med 2001;15:547–552.

13. Melendez LD, Twedt DC, Wright M. Suspected doxycycline-in-duced esophagitis with esophageal stricture formation in three cats. Feline Pract 2000;28(2):10–12.

14. Melendez LD, Twedt DC, Weyrauch EA, et al. Conservative therapy using balloon dilation for intramural, inflammatory esophageal strictures in dogs and cats: a retrospective study of 23 cases (1987–1997). Eur J Comp Gastroenterol 1998;3:31–36.

15. Harai BH, Johnson SE, Sherding RG. Endoscopically guided bal-loon dilatation of benign esophageal strictures in 6 cats and 7 dogs. J Vet Intern Med 1995;9:332–335.

16. Hardie EM, Greene RT, Ford RB, et al. Balloon dilatation for treatment of esophageal stricture: a case report. J Am Anim Hosp Assoc 1987;23:547–550.

17. Sooy TE, Adams WM, Pitts RP, et al. Balloon catheter dilata-tion of alimentary tract strictures in the dog and cat. Radiology 1987;28:131–137.

18. Papazoglou LG, Patsikas M, Rallis T, et al. Hiatal hernia with esophageal stricture in a cat. Feline Pract 2000;28(3):10–14.

19. Galatos AD, Rallis T, Raptopoulos D. Post anaesthetic oe-sophageal stricture formation in three cats. J Small Anim Pract 1994;35:638–642.

20. Wilson DV, Walshaw R. Postanesthetic esophageal dysfunction in 13 dogs. J Am Anim Hosp Assoc 2004;40:455–460.

21. Pearson H, Darke PG, Gibbs C, et al. Reflux oesophagitis and stricture formation after anaesthesia: a review of seven cases in dogs and cats. J Small Anim Pract 1978;19:507–519.

22. Grier RL. Esophageal disease as a result of improper patient po-sitioning. J Small Anim Pract 1975;4:4–6.

23. Harvey HJ. Iatrogenic esophageal stricture in the dog. J Am Vet Med Assoc 1975;166:1100–1102.

24. Jergens A. Diseases of the esophagus. In: Ettinger S, Feldman E, eds. Textbook of veterinary internal medicine. 6th ed. St Louis: Elsevier Saunders, 2005;1298–1309.

25. Pereira-Lima JC, Ramires RP, Zamin I Jr, et al. Endoscopic dila-tion of benign esophageal strictures: report on 1043 procedures. Am J Gastroenterol 1999;94:1497–1501.

26. Scolapio JS, Pasha TM, Gostout CJ, et al. A randomized prospec-tive study comparing rigid to balloon dilators for benign esoph-ageal strictures and rings. Gastrointest Endosc 1999;50:13–17.

27. Hernandez LV, Jacobson JW, Harris MS. Comparison among the perforation rates of Maloney, balloon, and savary dilation of esophageal strictures. Gastrointest Endosc 2000;51:460–462.

28. Dent J, El-Serag HB, Wallander MA, et al. Epidemiology of gastro-oe-sophageal reflux disease: a systematic review. Gut 2005;54:710–717.

29. Galatos AD, Raptopoulos D. Gastro-oesophageal reflux during anaesthesia in the dog: the effect of age, positioning and type of surgical procedure. Vet Rec 1995;137:513–516.

30. Wilson DV, Boruta DT, Evans AT. Influence of halothane, isoflu-rane, and sevoflurane on gastroesophageal reflux during anes-thesia in dogs. Am J Vet Res 2006;67:1821–1825.

31. Jaspersen D. Drug-induced oesophageal disorders: pathogenesis, in-cidence, prevention and management. Drug Saf 2000;22:237–249.

32. Beatty JA, Swift N, Foster DJ, et al. Suspected clindamycin-as-sociated oesophageal injury in cats: five cases. J Feline Med Surg 2006;8:412–419.

33. Emami M, Haghighi M, Esmaeili A. Esophagitis caused by cip-rofloxacin; a case report and review of the literature. Govaresh 2004;9:272–276.

34. Gould PC, Bartolomeo RS, Sklarek HM. Esophageal ulceration associated with oral penicillin in Marfan’s syndrome. N Y State J Med 1985;85:199–200.

35. Barrs VR, Beatty JA, Tisdall PL, et al. Intestinal obstruction by trichobezoars in five cats. J Feline Med Surg 1999;1:199–207.

36. Said A, Brust DJ, Gaumnitz EA, et al. Predictors of early re-currence of benign esophageal strictures. Am J Gastroenterol 2003;98:1252–1256.