Risk factors for refractory enterocutaneous fistula following button jejunostomy removal and its treatment using a novel extraperitoneal approach in patients with oesophageal cancer: a retrospective cohort study

Enterocutaneous fistula after removal of the jejunostomy tube leads to multiple problems, such as cosmetic problems, decreased quality of life, electrolyte imbalances, infectious complications, and increased medical costs. However, the risk factors for refractory enterocutaneous fistula (REF) after button jejunostomy removal remain unclear. Therefore, in this study, we assessed the risk factors for REF after button jejunostomy removal in patients with oesophageal cancer and reported the surgical outcomes of the novel extraperitoneal approach (EPA) for REF closure. This retrospective cohort study included 47 patients who underwent button jejunostomy removal after oesophagectomy for oesophageal cancer. We assessed the risk factors for REF in these patients and reported the surgical outcomes of the novel EPA for REF closure at the International University of Health and Welfare Hospital between March 2013 and October 2021. The primary endpoint was defined as the occurrence of REF after removal of the button jejunostomy, which was assessed using a maintained database. The risk factors and outcomes of the EPA for REF closure were retrospectively analysed. REFs occurred in 15 (31.9%) patients. In the univariate analysis, REF was significantly more common in patients with albumin level < 4.0 g/dL (p = 0.026), duration > 12 months for button jejunostomy removal (p = 0.003), and with a fistula < 15.0 mm (p = 0.002). The multivariate analysis revealed that a duration > 12 months for button jejunostomy removal (odds ratio [OR]: 7.15; 95% confidence interval [CI]: 1.38–36.8; p = 0.019) and fistula < 15.0 mm (OR: 8.08; 95% CI: 1.50–43.6; p = 0.002) were independent risk factors for REF. EPA for REF closure was performed in 15 patients. The technical success rate of EPA was 88.2%. Of the 15 EPA procedures, fistula closure was achieved in 12 (80.0%). The complications of EPA (11.7%) were major leakages (n = 3) and for two of them, EPA procedure was re-performed, and closure of the fistula was finally achieved. This study suggested that duration > 12 months for button jejunostomy removal and fistula < 15.0 mm are the independent risk factors for REF after button jejunostomy removal. EPA for REF closure is a novel, simple, and useful surgical option for patients with REF after oesophagectomy.