Intraoperative presentation of an undiagnosed tracheoesophageal fistula in an adult without history of abdominal or thoracic surgery

Abstract

Tracheoesophageal fistula is uncommon in adults but can cause devastating aspiration events. Herein, we report a unique case of a tracheoesophageal fistula in an adult that presented intraoperatively. The patient did not have any prior history of abdominal or thoracic surgery and was not intubated for a prolonged period of time. The diagnosis, subsequent hospital course, and recommendations for early recognition of this rare condition are discussed.

Pulmonary aspiration events in the perioperative environment can be devastating and are associated with high morbidity and mortality.^1,2 Presence of a tracheoesophageal fistula (TEF) is a rare cause of aspiration events and is uncommon in adults.³ Malignancy is the most common cause of TEF in adults, with less common causes including trauma, corrosive ingestion, congenital, and infection (e.g., tuberculosis).^4–7 Iatrogenic causes also exist, most often related to endotracheal intubation and cuff-related tracheal injuries. There are few documented cases of TEF in adults discovered intraoperatively.

CASE DESCRIPTION

A 41-year-old woman with type 2 diabetes, end-stage renal disease on dialysis, and necrotizing soft tissue infection of her foot that extended to her groin presented to the operating room for irrigation and debridement of her left lower extremity and sacral wound. She had been on supplemental tube feeds via a Dobhoff tube due to poor oral intake and malnutrition. In the operating room, she was noted to have a mild nonproductive cough but otherwise normal lung exam and x-ray findings. She had pulse oximetry readings of 100% while breathing air. Due to delays in the operating room schedule, she had been made nil per os for over 24 hours before surgery, including tube feeds. The case began uneventfully with intravenous induction and intubation with a 6.5 mm endotracheal tube (ETT) via direct laryngoscopy without difficulty. Halfway through the case, the patient was turned from supine to right lateral position to access her sacral eschar.

At the end of the surgery, the end-tidal carbon dioxide waveform on the ventilator was noted to have changed; the phase III alveolar plateau and the phase 0 inspiratory downstroke were slanted toward zero. A soft gurgling sound was also noted with each respiratory cycle. An additional 2 mL of air was applied to the cuff, but the leak persisted. As the patient was about to be returned to supine position, bilious material was noted to be passing through the ETT. She was immediately turned supine, and then 50 to 100 mL of bilious material passed directly into the ETT and the ventilator circuit. The ETT was immediately removed, and the patient’s oropharynx was suctioned thoroughly. She was bag-masked until a new circuit could be applied to the machine. Due to low oxygen saturation, she was reintubated with a 7.0 mm ETT for airway protection and concern for aspiration. A gurgling sound was again heard around her airway, which resolved with an additional 2 mL of air to the cuff. A nasogastric tube was inserted to decompress her stomach. She was transported to the surgical intensive care unit (ICU) in stable condition.

Postoperatively, the ICU team noted a large leak from the ETT, subsequently exchanged the ETT via bougie, and confirmed the position of the cuff below the vocal cords under GlideScope visualization. Despite this, the leak persisted. Additionally, whenever the nasogastric tube was placed on suction, a significant discrepancy between delivered tidal volumes was noted on the ventilator machine. At this time a presumed diagnosis of TEF was made.

The patient underwent a bedside bronchoscopy in the ICU, during which no TEF was found below the level of the ETT cuff. An esophagram was equivocal. Computed tomography of the chest revealed a fistula at 10 mm cephalad to the origin of the left mainstem bronchus. She then underwent esophagogastroduodenoscopy with the gastrointestinal team, during which insufflation of the esophagus and stomach resulted in high end-tidal carbon dioxide readings—a finding that can be sensitive for TEF. A linear ulceration was discovered along the anterior esophageal wall containing an apparent fistula tract into the trachea (Figure 1a). The presence of the fistula was confirmed with contrast injection under fluoroscopy before a stent was deployed and sutured in place (Figure 1b). Following the procedure, the patient tolerated ice chips, and her oral feeds were restarted. However, she suffered a massive aspiration event the day after the stent was placed and required reintubation. She was placed under do-not-resuscitate status by her family members and was extubated to comfort measures. She expired soon after extubation.

Figure 1.

Esophagogastroduodenoscopy images showing (a) a linear ulceration along the anterior esophageal wall (white arrow) and the tracheoesophageal fistula (red arrow) and (b) a stent placed along the esophageal ulceration, covering the tracheoesophageal fistula.

DISCUSSION

TEFs are rare in adults. There are few case reports of TEF diagnosed or discovered intraoperatively,^5,8 and most are in pediatric patients^6,9 or in patients with a history of abdominal or intrathoracic surgery.^4,8,9 Intraoperatively, most patients with TEF present with an air leak around the ETT despite appropriate position of the ETT cuff below the vocal cords, gastric distention with positive pressure ventilation, or loss of ventilator volume with gastric suction. Diagnosing a TEF intraoperatively requires a high degree of suspicion, as there are many other causes of air leak around the ETT.^4,10 Fiber-optic bronchoscopy through the ETT may identify a fistula past the level of the ETT cuff, but the definitive diagnosis will require endoscopy of the esophagus. Management of a known TEF perioperatively includes several steps: preoperatively, optimizing the patient’s nutritional status and obtaining routine labs and a type and cross; intraoperatively, obtaining an arterial line for close hemodynamic monitoring, ensuring the ETT cuff is positioned distal to the level of the TEF, and placing a Fogarty catheter prior to intubation or one-lung ventilation if necessary; and postoperatively, evaluating timing of extubation based on patient status and beginning enteral nutrition via a feeding tube distal to the anastomosis.¹¹

In our case, the patient had an undiagnosed TEF that was unmasked during surgery. While she did present with a mild cough, there were few other signs of TEF prior to the surgery. Potential iatrogenic etiologies include ulceration from the Dobhoff tube and pressure injury from multiple endotracheal intubations. Due to the presence of a linear ulceration around the fistula for this patient, the more likely etiology is the nasogastric tube ulcerating the esophageal wall into the trachea. The patient had been chronically ill and had poor nutrition status, which likely contributed to poor tissue integrity within the esophagus and trachea that ultimately predisposed to the formation of a fistula.

For patients, especially those with malnutrition, who have had a chronic nasogastric tube, chronic ETT, or multiple intubations within a short period of time (<3 months), we recommend consideration of the possibility of the presence of a TEF. Consider a diagnosis of TEF if a leak is noted around the ETT despite adding additional air to the pilot balloon, perform a bronchoscopy exam if possible to identify a fistula, and have a consultation for formal endoscopic evaluation if a high degree of concern persists, as well as for consideration of therapeutic options.

DISCLOSURE STATEMENT

The authors report no funding or conflicts of interest. Consent was obtained from the patient for publication of this case.