ESOPHAGEAL LEAK MANAGEMENT USING NASOCAVITY DRAIN AND BANDED FENESTRATED STENT

INTRODUCTION
Leakage of esophageal content through esophageal perforation can cause abscesses around the esophagus or pleural space. These abscesses are challenging to manage, mainly if the leak is in the proximal esophagus, where stent placement is challenging due to limited landing space. The dependent undrained part of the abscess typically ends up draining into the respiratory tract or pleural space, leading to complicated fistulas.

CLINICAL PRESENTATION
A 40-year-old male presented to the ER with difficulty swallowing. His entire esophagus was narrowed down to 5mm on his radiologic imaging with an extrinsic compression (Image 1). Endoscopic examination revealed that the para-esophageal leak was a large infected mediastinal hematoma that was compressing the entire esophagus.

ENDOSCOPIC METHOD
We cut a suction tubing to create a cap for a pediatric upper endoscope to investigate the leak. After identifying the leak and irrigating the purulent material, we noted a 12 cm long and 2 cm wide cavity. With guidewire assistance, a 12 fr 60 cm locking loop was used as a nasocavity drain (Image 1). The endoscopy was repeated on day 10. Either the pressure from the infected liquified hematoma or the presence of a drain in the abscess cavity resulted in more than one fistula opening. However, the distal part of the cavity was not adequately drained.

Instead of doing a long septotomy, we passed a guidewire from the proximal leak in the esophagus into the abscess cavity to dissect and open the distalmost part of the abscess into the esophagus. A fenestrated ureteral stent was placed over this wire. Due to the limited space and friable tissue, endoscopic suturing was not feasible to secure the stent. Therefore, we banded the two pigtails of the ureteral stent together to secure the stent and to provide constant drainage of the para-esophageal space. As shown in Image 2, the fenestrated ureteral stent exits the esophageal lumen at the proximal leak site, traversing the abscess, reentering the esophagus, and then the pigtails are secured with bands.

The patient remained asymptomatic, and the para-esophageal cavity resolved at seven weeks. The bands were cut, and the drain was removed.

CONCLUSION
This approach demonstrates a two-step approach. The first step is to extrinsically drain the infected and liquified hematoma with a nasocavity drain, and the second step is to resolve the abscess cavity by an intentional dependent opening and a banded fenestrated stent as an alternative to long septotomy.