Endotherapy of Upper GI Anastomotic Leaks Using a Modified Tygon Plastic Stent: Results of a Large Single Center Series

Abstract

Background: Upper gastrointestinal (GI) anastomotic leaks are associated with significant morbidity and mortality. The risks of repeat surgery are high and further compounded by the possibility of poor anastomotic healing. Endoscopic treatment with self expandable stents (SES) has been reported. SES are costly, and in the absence of associated luminal strictures, there is a high risk of stent migration; in the event of successful placement, SES especially metallic stents may be difficult to remove after fistula resolution. Aim: To examine the feasibility of a modified Tygon plastic stent for the treatment of postoperative upper GI anastomotic leaks. Methods: The clinical data of consecutive patients with upper GI anastomic leaks who were referred for endoscopic treatment with a modified Tygon plastic stent during the period from November 1997 to June 2007 were reviewed. Tygon plastic stents with a diameter ranging from 9-14 mm were individually tailored according to the location of the leaks. After endoscopic placement of a guidewire, the stent was inserted over a bougie without fluoroscopic monitoring. To prevent distal migration, a 7Fr Teflon catheter was attached to the proximal end of the stent and looped around the ear before being fixated with adhesive tape. Results: A total of 40 patients (80% male; mean age 63.5 years) with upper GI anastomotic leaks (post-esophagectomy: 22.5%; post-gastrectomy: 35%; post-esophago-gastrectomy: 37.5%; chronic anastomotic fistula: 5%) were treated with a modified Tygon plastic stent with a mean duration of 32 days (range: 2 - 209) after surgery. Forty percent of these patients had previous unsuccessful treatment with surgery (35%) or endoscopic fibrin glue injection (5%). Stent insertion was technically successful in all patients without any procedure related complications. After stent placement, 30% of patients died from other postoperative complications and severe underlying co-morbidities before resolution of the leaks. Successful resolution of leaks occurred in 96.4% of the remaining cases after stent placement, and the stents were subsequently removed easily without the need for endoscopy after a mean duration of 39 days (range: 17 - 130). Stent dislocation occurred in 15% of cases but successful reinsertion was possible in all instances. No complications occurred after stent placement. Conclusions: Endoscopic temporary placement of a modified Tygon plastic stent is a very effective and safe method for management of upper GI anastomotic leaks.