Introduction

A 7.75-year-old 9.1-kg sexually intact male Welsh Terrier was presented to the referring veterinary surgeon with dyspnea, multiple cervical dog bite wounds, and diffuse subcutaneous emphysema around the neck. Methadone^a (0.2 mg/kg, IV) and fluid therapy (90 mL/kg, IV, over a 1-hour period) were administered. Radiography revealed bilateral pneumothorax, worse on the left side. Thoracocentesis was performed with a butterfly needle, and 680 mL of air (left side, 600 mL; right side, 80 mL) was removed until negative pressure was achieved. There was a subsequent improvement in the severity of the dyspnea, but the dog was unable to maintain SpO₂ > 90% without flow-by oxygen supplementation. The dog was therefore referred for further investigation of suspected recurrent pneumothorax. A neck bandage was placed to prevent further contamination of the wounds during transportation and flow-by oxygen supplementation was provided during the journey. Owing to the dog’s poor respiratory status, the wounds were not assessed or treated prior to referral.

On arrival, the dog had signs of depression. Clinical examination revealed a respiratory rate of 20 breaths/min with mildly increased respiratory effort despite oxygen supplementation, heart rate of 100 beats/min with no pulse deficits, mean arterial blood pressure of 154 mm Hg (systolic, 186 mm Hg; diastolic, 119 mm Hg)^b, and rectal temperature of 36.3 °C. Air flow could be auscultated on both sides of the thorax. The hair over the thorax was clipped, and no penetrating wounds were identified.

The dog was anesthetized for CT assessment of the cervical and pulmonary trauma and to allow exploration of the neck wounds. The neck bandage was left in place during the CT examination to prevent further contamination of the wounds. Anesthesia was induced with midazolam^c (0.3 mg/kg) and propofol^d (3 mg/kg) to facilitate intubation with an 8-mm-diameter silicon cuffed ETT.^e Intubation was uncomplicated with no resistance to insertion of the ETT. Anesthesia was maintained with sevoflurane^f in oxygen^g (2 L/min) delivered by a circle breathing system and spontaneous ventilation. Intravenous fluid therapy was continued with Hartmann solution^h (5 mL/kg/h).

Computed tomography revealed moderate left-sided pneumothorax (in the dorsocaudal aspect), pneumomediastinum, and an interstitial-alveolar pattern in the ventral aspect of the left lung lobes. There was evidence of considerable hemorrhage, extensive deep facial subcutaneous emphysema, subcutaneous emphysema around the neck, and a marked gas-filled esophageal dilation; fluid was present in the caudal portion of the esophagus. There was no CT evidence of a tracheal tear, and the tip of the ETT was located at the thoracic inlet. A gastric foreign body, suspected to be a peach stone, was considered an incidental finding. Subsequent esophagoscopy confirmed the absence of an esophageal tear and allowed the gastric foreign body to be removed endoscopically. The neck was explored following removal of the bandage and clipping of the hair over the traumatized area. A 2-cm-long penetrating wound that extended into a large open pocket was identified on the left ventral aspect of the neck. Multiple smaller puncture wounds were also present in the cervical region with extensive subcutaneous emphysema, but no visible damage to underlying structures was identified. The wounds were lavaged with 2 L of warm saline (0.9% NaCl) solution by means of a pressurized bag, fluid administration set, and 21-gauge needle. The dog continued to ventilate spontaneously; the end-tidal carbon dioxide concentration and SpO₂ remained stable, and there was no increase in respiratory noise detected on auscultation. A tie-over dressing was placed and methadone (0.2 mg/kg) was administered IV before the dog was allowed to recover from anesthesia. Administration of sevoflurane was stopped, and the fresh gas flow was changed to oxygen (1 L/min) and medical air^g (1 L/min). After the dog had regained palpebral reflexes, it was disconnected from the breathing system and allowed to breathe room air (21% oxygen). The dog’s SpO₂ remained > 97% throughout this process, but its recovery from anesthesia was very slow. The dog was transferred to the intensive care unit for close observation. The ETT was left in place with the cuff inflated to minimize the risk of aspiration should regurgitation occur during transportation.

After the dog had been moved to the intensive care unit, the ETT cuff was deflated. However, the emphysema around the neck appeared to be increasing and the SpO₂ rapidly decreased to 82%. The Fio₂ increased to 100% after reattachment of the ETT to a breathing system and oxygen supply. The ETT cuff was reinflated, and the dog’s SpO₂ improved to > 90%. Sevoflurane administration was restarted to facilitate thoracic radiography. There was no radiographic evidence of recurrence of the pneumothorax or notable worsening of the contusions. The ETT cuff was visible at the level of the large penetrating cervical wound and rostral to its previous position. In the absence of radiographic signs of deteriorating lung injury, the sudden decrease in SpO₂ and worsening subcutaneous emphysema that coincided with deflation and movement of the ETT cuff suggested the presence of a tracheal tear. For rapid confirmation, without the requirement for more advanced imaging, the oxygen concentration at the skin wound was investigated with an oxygen analyzer.ⁱ The oxygen analyzer was calibrated based on the manufacturer’s recommendations with an oxygen flow of 2 L/min (100% oxygen) and then allowed to stabilize to room air (20.2% oxygen). After stabilization of the Fio₂ at 100%, the dog was disconnected from the breathing system, and the analyzer was placed at the end of the ETT. The oxygen reading increased to > 60%, thereby confirming that the gases exhaled by the dog contained a higher oxygen concentration than room air. The dog was then reconnected to the breathing system and the analyzer was allowed to restabilized to room air (20.2% oxygen). The cuff of the ETT was then deflated, and the ETT was pulled back so the distal end was rostral to the penetrating wound. The analyzer was then placed at the edge of the wound, and the measured oxygen concentration rapidly increased to 71%, a value similar to that measured previously for the expired gases (Supplementary Video 1). This high oxygen concentration confirmed that the gas leaking through the wound had a greater oxygen concentration than room air, confirming the presence of a tracheal tear that communicated with the cervical wound.

The dog was reintubated with the ETT, which was advanced to ensure that the inflated cuff was caudal to the penetrating wound. The oxygen analyzer was reapplied at the edge of the wound, and an oxygen concentration similar to room air (20.2% oxygen) was recorded, confirming that oxygen-rich gases were no longer leaking from the trachea and the ETT cuff had sealed the leak. The ETT was then disconnected from the breathing circuit, and the dog was allowed to breathe room air for 5 minutes. The dog’s SpO₂ was maintained at > 97% during this time, which confirmed that the cause of the acute oxygen desaturation was related to the tracheal tear. With hindsight, we could have advanced the ETT and monitored the response when the acute oxygen desaturation first occurred; a detectable improvement in SpO₂ would have been indicative of a tracheal defect, although further confirmation would have been required.

Following discussion with the owners, the dog was prepared for surgery, and a ventral midline approach used to allow surgical exploration of the tracheal wound. Anesthesia was maintained with sevoflurane in 1 L of oxygen/min, with continuous rate infusions of ketamine^j (10 μg/kg/min) and fentanyl^k (0.2 μg/kg/min). A tear was identified between the eighth and ninth tracheal rings, which extended completely through the mucosa along 50% of the tracheal circumference. Simple interrupted sutures of 4-0 polydioxanone^l suture material were preplaced around the tracheal rings on either side of the mucosal defect, and the ETT was moved to ensure the sutures had not penetrated it. The preplaced sutures were then tied, and a left sternothyroideus muscle flap was sutured over the repair with 4-0 polydioxanone suture material in a simple interrupted pattern for additional protection. The surgical site was lavaged with sterile saline solution, and no air leakage was identified on submersion of the trachea in the solution. The sternohyoideus and platysma muscles were closed with simple continuous 4-0 polydioxanone sutures, and the skin was closed with poliglecaprone 25^m sutures placed in the subcutaneous and intradermal layers. At the end of the surgery and with the ETT retracted rostrally, the oxygen analyzer was placed at the edge of the penetrating wound; the oxygen concentration at that location was not greater than that of room air. The dog slowly recovered from anesthesia, maintaining its SpO₂ > 97% when breathing room air.

Discussion

Bite wound trauma is a common emergency in veterinary practice, with head and cervical injuries representing 20% to 29% of canine bite wound trauma cases.^1,2 In 1 study,² dogs with head and cervical bite wound trauma had a mortality rate of 22%, compared with a rate of 7% among all dogs with bite wound trauma. Jordan et al³ reported that in 56 canine and feline neck trauma cases, 9 involved damage to the airway.

Because of the shearing, tensile, and compressive forces exerted during a bite attack, damage to underlying structures can also occur in addition to visible wounds.⁴ In the neck region where the skin and subcutaneous tissue are very mobile, superficial wounds may appear small and minor but the damage to the underlying structures can be extensive.⁵ Damage to vital structures including the trachea, esophagus, major blood vessels, vagosympathetic nerves, and glandular tissues can occur as a result of penetrating wounds,³ and the superficial location of those structures may explain the high mortality rate associated with cervical bite wounds. Airway trauma is potentially catastrophic; therefore, rapid diagnosis and treatment of airway damage is extremely important and is associated with better patient outcomes.^3,6

Various methods for diagnosis of tracheal tears in humans and other animals have been reported. Regardless of the method used to confirm the presence of a suspected tracheal tear in a patient, a detailed history should be obtained and a thorough clinical examination should be performed. Synchronization of a dynamic air leak through a communicating wound with ventilation is pathognomonic for an airway defect⁷; however, this is not always apparent.⁸ Subcutaneous emphysema may develop as a result of airway trauma because air can leak through a tracheal tear into the subcutaneous tissues. Because tracheal tears are potentially life threatening, the presence of cervical subcutaneous emphysema indicates a need for immediate further investigation.^3,9 Other causes of subcutaneous emphysema include air entering between the subcutaneous tissues through the penetrating wound itself, air leaking from the thoracic cavity, and air leaking from esophageal tears.¹⁰ Concurrent pneumothorax and pneumomediastinum are often also present with traumatic tears¹¹; thus, thoracic auscultation of dogs with cervical subcutaneous emphysema is always warranted. Stridor or stertor may also be evident if the upper airway is narrowed¹¹ or damaged. Dyspnea is often but not always associated with tracheal tears. The phase of respiration in relation to dyspnea can indicate the anatomic site of trauma. Inspiratory dyspnea is usually associated with extrathoracic tracheal injury and expiratory dyspnea is usually associated with intrathoracic tracheal injury.¹¹ This pattern arises because the negative pressure generated during inspiration can worsen a dynamic extrathoracic obstruction but will open the intrathoracic airways. Conversely, during expiration, relative positive pressure is generated, which can collapse a dynamic intrathoracic obstruction but push open an extrathoracic obstruction.¹² However, in trauma cases, this concept is often complicated by confounding factors, such as pneumothorax or lung contusions. The dog of the present report was anesthetized with a neck bandage in place; hence, the neck was not examined in the initial clinical examination. Furthermore, the ETT had sealed the airway before the wounds were assessed. If the neck bandage had been removed prior to induction of anesthesia, clinical examination of the dog may have revealed a dynamic leak and worsening subcutaneous emphysema, indicating the presence of the tracheal tear.

Radiography rarely provides a specific diagnosis of a tracheal tear unless regional interruptions in the trachea are identified,^13,14 but findings associated with airway damage include peritracheal air accumulation, subcutaneous or deep fascial emphysema, and pneumomediastinum.¹⁴ Peritracheal air accumulation was not obvious on the radiographs obtained by the referring veterinarian. However, the dog had extensive deep facial and neck subcutaneous emphysema prior to anesthesia, and a tracheal tear was considered even though it was not specifically identified during clinical examination. In small animals with polytrauma, CT is often the diagnostic method of choice because it provides a large amount of information regarding multiple body systems in a short amount of time^6,9; hence, CT examination of the dog of the present report was elected. The diagnosis of tracheal tears on the basis of CT findings, such as discontinuity of the tracheal wall and deformity of tracheal cartilages,¹⁵ has been reported.^6,15,16 Additional associated findings include the presence of peritracheal air, subcutaneous emphysema, pneumomediastinum, and pneumothorax,¹⁵ although these features are not pathognomonic for tracheal tears. In the case described in the present report, the trachea appeared intact and uniform and a tracheal tear was initially deemed unlikely, despite the presence of such associated CT findings. On later reexamination of the CT images, the tracheal tear was still not identifiable, probably because the ETT was maintaining the continuity of the trachea at the time of CT. In human medicine, bronchoscopy remains the gold standard for confirmation of tracheal and bronchial tears,^17–19 although small tears may not always be identifiable.^17,18 To allow assessment of the entire length of the trachea, extubation is always required; however, this maneuver complicates anesthesia in an animal with respiratory compromise. In addition, bronchoscopy can be time consuming and requires expensive equipment and specialist expertise. Some argue that bronchoscopy is only appropriate in stable patients and when surgical exploration is unlikely to be performed.³ However, in some emergency cases, especially those involving large tracheal defects and complete ruptures, bronchoscopy-guided intubation may be required^20,21; moreover, bronchoscopy is the only procedure that can reliably exclude tracheal damage, which is important to establish in emergency trauma cases.²¹ For the dog of the present report, bronchoscopy should probably have been performed at the time of esophagoscopy and may have identified the need for surgery sooner. The decision not to perform bronchoscopy was based on the interpretation of the CT images and the desire to minimize anesthesia time.

In the case described in the present report, use of an oxygen analyzer rapidly confirmed the tracheal tear and the indication for surgical intervention. The analyzer consisted of a noble metal cathode and a base metal (lead) anode surrounded by a weak acid electrolyte solution. A plastic gas-permeable membrane separated the cathode from the gas under investigation. Because oxygen is consumed at the cathode, it creates a potential difference between the cathode and anode that is proportional to the oxygen tension in the gas.²² When placed at the edge of a penetrating wound that is leaking oxygen-rich gases, the oxygen analyzer reading will increase, compared with the reading for room air. Oxygen-rich gas would be present at the edge of a penetrating wound only if the wound is communicating with gas in the trachea and the patient is breathing in a gas mixture with a high Fio₂. This test provided a quick and simple diagnosis and, for this dog, eliminated the requirement for further advanced imaging.

One limitation of the oxygen analyzer technique for tracheal tear detection is that the patient must be breathing a gas mixture with a high Fio₂; ideally, endotracheal intubation or placement of a laryngeal mask is required. In theory, a tight-fitting face mask with provision of 100% oxygen may be sufficient, assuming a lack of oxygen leakage around the mask can be confirmed. As such, this could possibly be performed in a conscious animal. However, this technique may not identify all tracheal tears because some may be present without air leakage through a communicating wound or leakage only into the subcutaneous tissues. Bronchoscopy still remains the gold standard method of diagnosis of tracheal tears, although in many polytrauma cases, CT is often indicated because of multiple injuries.

As described, the use of an oxygen analyzer as part of the diagnostic investigation of a suspected tracheal tear in a dog with cervical trauma was easy, noninvasive, and rapidly applied. This technique does not eliminate the need for a thorough clinical examination, advanced imaging, and bronchoscopy but would be useful for the confirmation of a tracheal tear that communicates with a skin wound, especially in situations where financial limitations or lack of equipment availability preclude undertaking advanced imaging or bronchoscopy.

Supplementary Materials

Supplementary materials are available online at: avmajournals.avma.org/doi/suppl/10.2460/javma.259.8.880.