An esophageal fistula is an abnormal communication between the esophageal lumen and surrounding structures, most commonly the respiratory system. Fistulas resulting in esophagoaortic, tracheoesophageal, or bronchoesophageal communications are uncommon despite the intimate relationship between the esophagus and the aorta, trachea, and bronchi. The most commonly reported esophageal fistula in dogs is the bronchoesophageal fistula (BEF). All reports of dogs with BEFs have involved small breeds, with miniature poodles and terriers being overrepresented.
Congenital and acquired BEFs have been described, with the latter being far more common. In humans, congenital BEFs have been associated with an uncoordinated separation of the esophagus from the respiratory tract during embryologic development, resulting in a persistent attachment between the two. Alternatively, BEFs could result from an intrauterine infection causing the embryonic bronchus and esophagus to adhere. Acquired BEFs are most commonly sequelae of esophageal perforation due to foreign bodies, chronic irritation, or, less frequently, pulmonary abscesses. The most frequently reported cause of acquired BEF in dogs is trauma caused by a retained esophageal foreign body, usually a bone. Foreign bodies may also penetrate the esophageal wall and establish a fistula with the trachea, pulmonary parenchyma, or skin. The greater incidence of esophageal fistulas with bronchi versus other sites has been attributed to the fact that most obstructions of the thoracic esophagus occur caudal to the heart.
It is believed that the pathogenesis of foreign body–induced BEF begins with esophageal wall necrosis and perforation, followed by leakage of esophageal contents into adjacent tissues. The necrotic reaction progresses, and an esophageal traction diverticulum results from the inflammatory reaction between the esophagus and bronchi. It is also possible that an esophageal diverticulum, either congenital or acquired, may be present before the acquired BEF forms.8 Healing of the foreign body–induced lesion eventually leads to development of a communicating tract and continuous airway contamination with esophageal contents. Little published information regarding the length of time required for BEF formation exists; however, some authors speculate that fistula formation may occur within several days after foreign body–induced esophageal damage.
BEFs allow ingested material to pass into the bronchi and lungs and may allow passage of pulmonary secretions into the digestive tract. Most canine BEFs due to esophageal foreign bodies connect the esophagus with either the right caudal lung lobe bronchus or the middle lung lobe bronchus. The right caudal lung lobe bronchus is reportedly the most frequently involved bronchus. Other causes of BEF formation include trauma, neoplasia, bronchial foreign bodies, and peri-esophageal inflammation. BEFs are rare in dogs but should always be considered when an esophageal diverticulum is found or eating/drinking is associated with coughing. Because of their chronic nature, insidious onset, and nonspecific clinical signs, BEFs may go undiagnosed. Human patients have reportedly reached adulthood before the condition was recognized.
History of an esophageal foreign body and a confirmed diverticulum are present in most cases of acquired BEF in dogs. The physical examination may reveal coughing, dyspnea, respiratory crackles over affected lung regions, hemoptysis, anorexia, depression, weight loss, regurgitation, and dysphagia. Coughing associated with drinking liquids, a sign frequently associated with BEFs, may take several days to develop. Coughing associated with eating or drinking strongly suggests a communication between the esophagus and tracheobronchial tree. Other reported complications include pyothorax, septicemia, and abdominal distention secondary to dyspnea and aerophagia. Excessive production of mucus shed via the nose and mouth may be an important early sign of the defect.
Death may result from drowning subsequent to accumulation of secretions or from pneumonia due to aspiration of gastric and esophageal contents. Large fistulas directed caudally or ventrally from the esophagus produce a more consistent cough than small fistulas directed cranially or dorsally, most likely as the result of gravity. Systemic signs in dogs are associated with mediastinitis, aspiration pneumonia, or bronchopneumonia. Respiratory signs may be relatively mild, resulting in a marked delay in the diagnosis of a congenital lesion. In humans, coughing associated with swallowing liquids may be brought on by a change of posture, such as lying on the back or side. In some cases, dysphagia may be the predominant sign.
Diagnosis of BEF is based on presenting clinical signs together with other diagnostic tests.
A complete blood count may reveal inflammatory changes associated with chronic pneumonia. Survey thoracic radiography may reveal radiopaque foreign bodies in the esophagus or bronchus, pulmonary consolidation, pleural fluid accumulation, or localized interstitial, alveolar, or bronchial lung patterns. Pleural fluid may accumulate secondary to localized pulmonary infection or as an extension of the inflammatory reaction around the fistula.
A BEF should be suspected in a young animal with recurrent aspiration pneumonia; a focal, recurrent lung opacity; or localized pneumonia associated with coughing after eating or (especially) drinking. With the exception of the case presented here, characteristic lung patterns typically involve disease localized to the right caudal and/or middle lobes. In the case in this report, disease was localized to the accessory lobe, another possible site of aspiration pneumonia.
Contrast esophagography, endoscopy, or histopathology is required for definitive diagnosis of communication between the esophagus and the airway, with contrast esophagography considered the gold standard. Most authors recommend performing an esophageal contrast study when a BEF is suspected or when an esophageal diverticulum is identified by other diagnostic procedures. Even if clinical signs are not associated with swallowing, a barium esophageal contrast study should be part of evaluating chronic pulmonary disease of unknown etiology (Figure 7 and Figure 8). Iodinated contrast agents should be avoided because they are hyperosmolar and result in pulmonary edema.
Oral iodinated contrast agents are also more irritating to bronchi and might elicit a cough, which would result in poor opacification of the fistula and/or bronchi.8 Barium solution is preferred if communication with the bronchi is suspected because of its low cost and its nonreactivity in airways. A thin mixture of barium should be used (20% to 30% weight/volume) because it fills small fistulas more efficiently than thick barium mixed with food. However, small fistulas may be more difficult to demonstrate. Fluoroscopy with some form of video recording is helpful in diagnosing small fistulas or fistulas directed cranially or dorsally from the esophagus. In these cases, the contrast medium may only momentarily fill the bronchus. Multiple recumbent positions may be required to facilitate gravitation of the contrast medium into or through the fistula.8 Fluoroscopy is also helpful in differentiating a fistula from aspiration when contrast medium is used. In humans, repeated esophageal contrast studies may not uniformly reveal the fistula even when it is strongly suspected.
Esophagoscopy and bronchoscopy can be attempted, but both require general anesthesia, which may be a problem in dogs with compromised respiratory function. Moreover, endoscopy may not allow clear identification of the origin of the fistulous tract. Bronchoscopy may allow visualization of the diseased lung lobe(s) and enables sample collection from affected areas for cytology and culture. The presence of Simonsiella spp is only significant if oral contamination of the catheter or fluid is avoided during transtracheal aspiration,15 as was seen in the case presented. Predictably, bacteria cultured from fluid obtained via transtracheal or bronchoalveolar lavage reflect their oral and esophageal sources. Bacteroides fragilis, Actinomyces spp, Enterococcus cloacae, Escherichia coli, Staphylococcus intermedius, and nonhemolytic and α-hemolytic streptococci have been reported.
If the patient succumbs to complications or is euthanized, the diagnosis of a BEF may be confirmed based on gross and histopathologic examination. In human medicine, the following criteria are generally accepted and used to confirm the congenital origin of a BEF5: (1) absence of past or present surrounding inflammation; (2) lack of adherent lymph nodes; and (3) presence of a definite mucosa and muscularis mucosa within the fistula.
Treatment consists of surgical correction of the fistulous tract. Anesthesia may present a challenge because the fistula makes ventilation difficult, and inhaled anesthetic escapes into the esophagus,21 which can result in tracheobronchial flooding with gastric contents. In human patients, this complication has been prevented by preoperative gastrostomy or endobronchial intubation. A dog with a BEF between the esophagus and right caudal lobar bronchus underwent intubation of the left main stem bronchus under fluoroscopic guidance to avoid loss of anesthetic gases and ventilatory problems when the fistula was being dissected.
Lobectomy may be necessary if extensive pulmonary lesions such as consolidation, abscessation, bronchopneumonia, fistula recurrence, and foreign material contained within the airways are present. Postoperatively, an esophageal contrast study may be performed to confirm normal esophageal motility and resolution of the BEF.
The prognosis ranges from good to guarded and depends on the preoperative condition of the patient, ability to resolve pulmonary infection, success of surgery, and presence of clinical complications such as esophageal trauma, thoracic involvement, pyothorax, septicemia, dehiscence, stricture, and pulmonary abscess. A good prognosis is associated with successful surgery and minimal clinical complications.
- Parker NR, Caywood DD. (1987) Surgical disease of the esophagus. Vet Clin North Am Small Anim Pract 17(2):333-358
- Caywood DD, Feeney DA. (1982) Acquired esophagobronchial fistula in a dog. JAAHA 18:590-594
- Guilford WG, Strombeck DR. (1996) Diseases of swallowing. In: Strombeck DR, Guilford WG, eds. Strombeck's Small Animal Gastroenterology. 3rd ed. Philadelphia: WB Saunders; pp:211-238
- Nawrocki MA, Mackin AJ, McLaughlin R, et al (2003) Fluoroscopic and endoscopic localization of an esophagobronchial fistula in a dog. JAAHA 36:257-261
- Jergens AE. (2005) Diseases of the esophagus. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine, 6th ed. St. Louis: Elsevier Saunders; pp:1298-1310
- Hedlund CS. Surgery of the esophagus. In: Fossum TW, ed. Small Animal Surgery. 2nd ed. St. Louis: Mosby; pp:307-337
- Johnson LR. (2005) Diseases of the small airways. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine. 6th ed. Philadelphia: Elsevier Saunders; pp:1233-1239
- Gualtieri M. (2001) Esophagoscopy. Vet Clin North Am Small Anim Pract 31(4):605-630
- Foor JC. (1974) A congenital bronchoesophageal fistula in a dog. Minnesota Vet 14:9-28
- Thrall DE (1973) Esophagobronchial fistula in a dog. Vet Radiol Ultrasound 14(1):22-25
- Braimbridge MV, Keith HI. (1965) Oesophago-bronchial fistula in the adult. Thorax 26:226-233
- Shaw DH. (2004) Bronchoesophageal fistulas. In: King LG, ed. Textbook of Respiratory Disease in Dogs and Cats. St. Louis: Elsevier Saunders; pp:397-399
- Fox SM, Allan FJ, Guilford WG, et al (1995) Broncho-oesophageal fistulas in two dogs. N Z Vet J 43:235-239
- Muir P, Bjorling DE. (1994) Successful surgical treatment of a broncho-oesophageal fistula in a cat. Vet Rec 134:475-476
- Nelson RJ, Benfield JR. (1970) Benign esophagobronchial fistula: a curable case of adult pulmonary suppuration. Arch Surg 100:685-688
- Basher AWP, Hogan PM, Hanna PE, et al (1991) Surgical treatment of a congenital bronchoesophageal fistula in a dog. JAVMA 199(4):479-482
- Kyles AE. (2002) Esophagus. In: Slatter DH, ed. Textbook of Small Animal Surgery. 3rd ed. Philadelphia: Elsevier Science; pp:573-592
- Busch DS, Noxon JO, Merkley DF. (1991) Bronchoesophageal fistula in a dog. Canine Pract 16(2):25-29