Complete endoscopic debridement combined with partial gastric wall resection successfully treated refractory esophago-gastric anastomotic fistula

A 59-year-old man presented with an esophagogastric anastomotic fistula following Ivor Lewis esophageal cancer resection. The Interventional Radiology Department treats patients with the new “three-tube” method, which involves the fluoroscopically guided transnasal placement of a sinus drainage tube, a nasogastric decompression tube, and a nasojejunal nutritional tube. However, after 6 months of treatment, his chest pain and fever had not improved, he was unable to eat orally, and pus was still coming out of the sinus drainage tube. Upper gastrointestinal radiography showed a fistula still present (Fig. 1a).

After he was transferred to our department, we decided to perform further treatment. Endoscopy showed plenty of pus in the upper gastrointestinal tract. After cleaning it up, we saw a large anastomotic fistula. Swollen mucosa covered the fistula and interfered with drainage, which was removed with a snare (Fig. 1b,c). A large amount of dense necrotic tissue in the fistula was removed by a hook knife (Fig. 1d,e). The gastric wall between fistula and gastric lumen was removed with a hook knife and a snare in order to open the fistula for adequate drainage (Fig. 1f). During the operation, coagulation forceps were used to stop the bleeding (Video S1). We placed a tube into the fistula, rinsed daily with 8000 mL of saline, a nasogastric tube for negative pressure drainage, and a nasojejunal tube for feeding. Two days later, endoscopy showed no pus in the fistula, and all tubes were removed. He was started on an oral liquid diet, and discharged.

Surveillance endoscopy after 1, 4, and 7 months (Fig. 2a–c) showed a good healing process. Then 17 months later, endoscopy and computed tomography showed complete healing of the fistula (Fig. 2d,e), and a weight gain of 9 kg during follow-up.

Overall, anastomotic fistula is a refractory disease, this case demonstrates that this method is safe and valid and deserves to be promoted.

Authors declare no conflict of interest for this article.

This work was supported by grants from Zhongyuan Talent Program (ZYYCYU202012113).

Approval of the research protocol by an Institutional Reviewer Board: N/A.

Informed Consent: Informed consent was obtained from the patient for the publication of their information and imaging.

Registry and the Registration No. of the study/trial: N/A.

Animal Studies: N/A.