Endoscopic treatment for post-operative fistulas of the upper digestive tract: FISTULENDO study

Aims Upper gastrointestinal cancers are common and come with a high mortality. Surgical treatment of these cancers is the cornerstone of curative care. One of the most feared surgical complications is the anastomotic fistula and its endoscopic management is a major challenge in clinical practice. The aim of this study was to evaluate the efficiency of endoscopic management of postoperative fistulas of the upper gastrointestinal tract.

Methods This was a prospective, multicenter, descriptive French study. Patients with an early fistula (< 30 days) after upper gastrointestinal surgery for cancer were included. Endoscopic management was performed based on an expert pre-determined algorithm. The primary outcome measure was clinical success (fistula healing), with no need for surgery. Fistula healing was defined as the absence of a cutaneous fistulization and CT collection 3 months after stent removal. The secondary objectives were: technical success, mortality, adverse effects, total duration of treatment, and length of hospital stay.

Results Seventy-four patients, 78.4% male, with a mean age of 64.9 years (+ /- 9.9) were included in the study. 35 patients (48.6%) had esophageal cancer, 30 patients (40.5%) had gastric cardia cancer and 9 patients (12.2%) had gastric cancer. 59 patients (79.7%) had an adenocarcinoma. 57 patients (77.0%) had a Lewis Santy procedure and 13 patients (17.6%) had a total gastrectomy. An abscess complicated the fistula in 23 patients (31.1%), with a mean size of 107.4 mm (+ /- 275.5), of which 54.2% were drained radiologically (53.8%) or surgically (46.2%). The type of anastomosis leakage was focal (< 2cm or<25% of the circumference) in 23 patients (33.8%), partial in 18 patients (26.5%), small (< 5mm) in 25 patients (36.8%) and complete in 2 patients (2.9%). Double pigtail plastic stents were placed in 13 patients (17.6%) and 24mm covered metal stents in 60 patients (81.1%). The technical success rate was 100%. The average rate of endoscopic procedures was 2.5 (1 – 5) per patient (total 182 procedures). Adverse events occurred during the procedure in 6 patients (3.3%) and post procedure in 12 patients (16.2%). Repeat surgery was needed in five patients (6.8%) due to failure of endoscopic treatment. The mean length of hospital stay was 33.9 days (+ /- 22.9). Overall mortality was 17.2% (10 patients) and 5 deaths occurred after the 3-month evaluation. No endoscopy-related death was noted. At 3-months, data was available in 65 patients (87.8%), with cutaneous fistulization in one patient and residual collection on CT in two patients. Clinical success was observed in 57 patients (87.7%), and 77.0% in intention to treat.

Conclusions In this French multicenter prospective study evaluating the endoscopic treatment of upper digestive fistulas after surgery for cancer, the clinical and technical success rates were 87.7% and 100% respectively, with an adverse event rate of 3.3%. These results are encouraging and must be confirmed by additional studies.