Tracing the track: From chronic cough to cervical fistula

Dear Editor,

Pulmonary cavitation is a frequent radiological finding with a broad differential diagnosis. Most common differential diagnoses are infectious aetiologies such as tuberculosis, necrotising bacterial pneumonia and invasive fungal disease.[1] Nevertheless, noninfectious and iatrogenic causes should be considered when initial investigations are inconclusive.

We report an unusual case of pulmonary cavitation resulting from a delayed tracheoesophageal fistula caused by erosion of cervical osteosynthesis material following anterior cervical discectomy and fusion (ACDF).

A 39-year-old man was referred for evaluation of a cavitary lung lesion [Figure 1] detected on chest computed tomography (CT). His medical history included juvenile rheumatoid arthritis and ACDF at C6–C7 performed 5 years earlier after a motor vehicle accident. He was a former smoker (30 pack-years) and was not taking chronic medication. The patient reported a 5-month history of productive cough, gastroesophageal reflux and dysphagia for liquids. He denied fever, weight loss, hemoptysis or tuberculosis exposure. Physical examination was unremarkable except for low body mass index (17.3 kg/m²).

Figure 1.

High-resolution computed tomography (axial view) of the chest showing RUL cavitation

Laboratory evaluation revealed mild leukocytosis, anaemia, thrombocytopenia and elevated inflammatory markers. Autoimmune screening was negative. Chest CT showed a thick-walled cavitary lesion in the right upper lobe measuring 6.5 cm, communicating with the corresponding bronchus and associated with fibrosis and traction bronchiectasis. Sputum cultures, including acid-fast bacilli testing, were negative. Initial flexible bronchoscopy demonstrated diffuse inflammatory changes without evidence of fistulisation.

Given persistent clinical suspicion, repeat CT imaging was performed and demonstrated air and soft-tissue thickening surrounding the cervical fixation material, extending into the superior mediastinum. A paravertebral collection established fistulous communication with the oesophagus and posterior tracheal wall [Figures 2 and 3], with distal extension to the right upper lobe bronchus. Repeat bronchoscopy identified a 0.8 cm tracheal fistula proximal to the carina, and upper gastrointestinal endoscopy confirmed an oesophageal fistula.

Figure 2.

High-resolution computed tomography of the chest (sagital view) showing fistulous tract.

Figure 3.

Reconstruction CT scan images showing displaced osteosynthesis material

Management was planned by a multidisciplinary team. Nutritional support with percutaneous endoscopic gastrostomy was initiated, followed by surgical removal of the cervical osteosynthetic material [Figure 4] and closure of the oesophageal fistula. The tracheal fistula was managed conservatively. At 3-month follow-up, complete resolution of the tracheal fistula was documented, oral feeding was restarted and a residual pulmonary cavity persisted under surveillance.

Figure 4.

Surgically removed osteosynthesis material

Delayed tracheoesophageal fistula after ACDF is rare and is thought to result from chronic erosion due to prolonged contact between fixation material and adjacent tissues, leading to pressure necrosis and foreign body reaction.[2,3] Although dysphagia is a common early postoperative complaint after ACDF, late-onset symptoms should raise suspicion for delayed complications.[4] Clinical manifestations of tracheal injury are often nonspecific, which may delay diagnosis. This case is notable because pulmonary cavitation was the initial radiological clue, emphasising the importance of maintaining a broad differential diagnosis when evaluating cavitary lung lesions.[1]

While surgical intervention is usually required for oesophageal fistula repair, selected small tracheal fistulae may heal with conservative management if carefully monitored, as demonstrated in this case.[5]

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.