Cutaneopulmonary Fistula in a Dog Caused by Migration of a Toothpick

Case Report

A 22-kg, male neutered, 6-year-old, mixed-breed dog was referred for evaluation of a chronic, nonhealing, draining tract located over the left cranial thoracic wall. The dog had an abscess form over that region approximately 1 year prior to presentation. The abscess was incised, explored, and a drain was left in place for a few days to maintain drainage. The abscess recurred within 3 weeks of this procedure. The abscess was explored and drained a second time, again with no resolution. The resultant wound from the second surgery formed a draining tract and had been treated with multiple antibiotics, enrofloxacin being the most recent prior to presentation. The dog had continued to eat and drink well and was showing no clinical signs of systemic infection, although an intermittent, moist cough had been noted by the owner.

On initial physical examination, a draining tract associated with the left thoracic wall was visualized. Moderate, generalized lymphadenopathy was also noted. The drainage from the wound was described grossly as purulent exudate. The remainder of the examination was unremarkable, including a normal ambient body temperature and unremarkable auscultation of the pulmonary fields.

Serum biochemistries revealed a hyperproteinemia (9.6 g/dL; reference range, 5.1 to 7.8 g/dL), hyperglobulinemia (6.8 g/dL; reference range, 2.3 to 4.5 g/dL), mild hyperglycemia (146 mg/dL; reference range, 60 to 125 g/dL), and decreased albumin/globulin ratio (0.4; reference range, 0.75 to 1.9). These findings were indicative of chronic antigenic stimulation. A complete blood count was within reference ranges.

Survey radiographs of the thorax revealed a mild alveolar pattern associated with the caudal aspect of the left cranial lung lobe. No evidence of pleural effusion was noted. Due to the chronicity of the fistula, surgical exploration was postponed in order to further define the wound. A fistulogram was completed to determine the extent of the draining tract and to evaluate for a foreign body within the tract. An 8-French Foley catheter was placed 3 cm into the fistula, and 10 mL of low osmolar, water-soluble, nonionic iodinated contrast material^a was injected into the stoma of the fistulous tract. No anesthesia was necessary to perform this study. Right lateral and ventrodorsal thoracic radiographs were taken. These radiographs revealed contrast within the left bronchial tree of the cranial and caudal lung lobes [Figure 1]. The contrast extended from the cutaneous surface through the thoracic wall, which confirmed a cutaneous-pulmonary fistula with a possible foreign body at the level of the costochondral junction, as evidenced by a thickened region of the contrast column visualized at the level of the fifth and sixth costochondral junction. The contrast was disseminated within the bronchial tree, which was probably a result of a mild cough noted during the study.

Surgical exploration of the fistula was indicated at that time. The dog was premedicated with buprenorphine (0.01 mg/kg body weight, intramuscularly [IM]) and acepromazine (0.05 mg/kg body weight, IM) 2 hours prior to surgery. Anesthesia was induced with ketamine (0.1 mg/kg body weight, intravenously [IV]) and diazepam (0.1 mg/kg body weight, IV) and was maintained with isoflurane in oxygen. During the procedure, the dog was mechanically ventilated. The dog received cefazolin (22 mg/kg body weight, IV) at the time of anesthetic induction and every 2 hours during surgery. The left thoracic region was clipped and aseptically prepared for surgery. A left fifth intercostal thoracotomy was performed. The fistulous tract was followed through the thoracic wall to the level of the costochondral junction. A fibrous structure (6 cm × 1 cm) was found at the level of the costochondral junction within the thoracic cavity. This structure was excised, and a deli toothpick was found within it. Mild thickening of the soft tissues surrounding the fifth costochondral junction was also present, most likely due to chronic inflammation. A small portion of the caudal aspect of the left cranial lobe was adhered to the fistulous region, and a partial lobectomy was performed using a thoracoabdominal stapling device,^b resulting in complete excision of the cutaneopulmonary communication. A sample for culture and sensitivity was collected during surgery by placing a culturette swap into the exposed fistula. Isolates from the tract included Escherichia coli and a Group D Enterococcus spp.; both were sensitive to multiple antibiotics, including ampicillin, cephalothin, and enrofloxacin. A chest tube was placed and secured at the time of surgical closure. The thoracic cavity was lavaged with warmed saline solution. The thoracic wall and skin were closed routinely. The thoracic cavity was evacuated of air through the chest tube. A Fentanyl patch^c (75 μg per hour transdermal) was placed prior to extubation.

Postoperatively, the dog recovered without complication. Pain management was maintained with morphine (0.5 mg/kg body weight, IM) every 4 hours for the first 24 hours. The chest tube was maintained for 36 hours, and minimal fluid and air were observed. Amoxicillin-clavulanic acid (22 mg/kg body weight, orally per os [PO]) q 12 hours was prescribed for 6 weeks. The dog was discharged from the hospital on day 3 postoperatively. The dog continued to improve clinically, with complete resolution of the tract seen during a follow-up examination 2 weeks later. No cough was noted throughout the recovery period.

Discussion

Foreign body migration is not an uncommon syndrome in dogs. In the veterinary and human literature, there are multiple accounts of foreign body migration.^1–13 Clinical outcomes range from small, cutaneous, inflammatory reactions to severe inflammation and bacterial infection of body cavities, organs, or both. Diagnosis and treatment of this condition can be challenging at times. Demonstrating the presence of a foreign body can be difficult, and successful treatment usually hinges on the complete excision of the object and the majority of diseased tissue associated with it.¹³⁸

In the dog, many different objects have been described as migrating foreign bodies, including wood fragments, grass awns, needles, and deli toothpicks.^127–9 This report is the first case, to the authors’ knowledge, of a toothpick causing a cutaneopulmonary fistula. In the human literature, there have been multiple reports of toothpick migration and toothpick injuries.^4–61011 In a study documenting all toothpick-related injuries in humans between the years 1979 and 1982, an estimated 8,176 injuries occurred every year, the majority occurring in younger children.⁵ These injuries have been documented to occur in multiple body regions, including the eyes, head, neck, the extremities, gastrointestinal tract, urinary system, and abdominal and thoracic cavities.^4–61011 In review of the veterinary literature, only two cases of toothpick-related injury could be found.² Based on the data collected within the human literature, it is probably safe to speculate that many more toothpick-related injuries occur in dogs than have been reported in the literature. Toothpicks are long, slender, hard, sharp, indigestible objects that have the potential to cause harm. In dogs, the most probable route of injury would be ingestion of the toothpick with subsequent migration. It is believed that esophageal migration was the initial route of migration injury in this case, although direct cutaneous migration or aspiration and subsequent migration cannot be ruled out. Fistulas that have been previously reported in association with a migrating foreign object include esophagoaortic, esophagotracheal, and esophagobronchial, with the latter being the most common in dogs.³

When attempting to diagnose a migrating foreign body, many modalities can be used. If peritonitis or pyothorax is present, survey radiographs, blood work, abdominocentesis/thoracocentesis, and surgical exploration may suffice. If a draining tract is present, one additional, simple diagnostic test that can be completed is contrast radiography. Sinography (i.e., injection of iodinated contrast into a sinus or draining tract) or fistulography (when the result indicates a cutaneocavitary communication) is a quick, inexpensive tool to help in identifying the presence of a foreign object. Historically, these types of radiographic studies have been used frequently in equine practice, but there has been sparse reporting of their use in small animal practice.² In a previous study, sensitivity for sinography diagnosing foreign bodies in small animals was reported as 87%.² This study also reported that in 44% of the animals examined, sinography demonstrated that the extent, the position of the draining tract, or both were different than expected on the basis of clinical signs and survey radiographs.² This emphasizes that valuable information can be produced by the use of contrast radiography. Other modalities that have been used to evaluate animals for foreign bodies include ultrasound, magnetic resonance imaging, and computed tomography.⁷ These modalities can be very useful, although they are not a convenient modality for initial evaluation because of their requirements of anesthesia, expertise, or expensive instrumentation. This case is an excellent example of how the use of fistulography, a diagnostic test available to any practitioner with radiography capability, can gain valuable information that can help with surgical planning and producing an increased probability of a favorable outcome. It is important to note that low osmolar, water-soluble, nonionic contrast agents are the agents of choice, and especially in this case where the agent came in direct contact with pulmonary tissue. Ionic agents are contraindicated in fistulograms that may enter the pulmonary parenchyma, because they can stimulate severe inflammatory reactions in pulmonary tissue, which can result in fatalities.

As with this case, the majority of draining tracts and chronic foreign bodies result in complete resolution of clinical signs, with complete excision of all diseased tissue and the foreign object. White and Lane report that 88%, and Lamb, et al., report that 85% of cases resolve with complete surgical excision of the tract and associated foreign body.²¹³ Possible reasons for recurrence following surgery would be inadequate removal of diseased tissue or the inability to remove a foreign body. Hence, it is well worth the effort to gain as much information as possible prior to surgical exploration.

Conclusion

This case demonstrated a rare type of foreign body presentation. With complete diagnostic workup including contrast radiography, a good understanding of the extent of lesion was obtained. With this information, appropriate surgical and postoperative plans were formed. This enabled the surgeon to completely excise all diseased tissue including the foreign body, which is imperative to complete healing in these types of cases. This dog, as expected, based on excision of both foreign body and diseased tissue, had complete resolution of the clinical signs within weeks of the surgery.