Introduction

Anastomotic leaks (AL) are a challenging complication after upper gastro-intestinal (UGI) surgery, with significant mortality and morbidity. Endoscopic vacuum therapy (EVT) has recently emerged as the mainstream tool for the management of UGI AL, both with commercially available devices and with in-house devices. Our group has recently (since July 2022) started to use bespoke, in-house made devices for EVT.

Aims & Methods

The aim of this study is to describe our experience with EVT for UGI AL in oncological patients undergoing surgery. We retrospectively collected and analysed data from all our patients who underwent EVT for AL after UGI oncological surgery, between July 2022 and December 2023 (18 months period). EVT was offered with bespoke, in house, hand-made negative pressure endoscopic devices connected to commercial, medical vacuum pumps. EVT was performed twice per week and initial response was assessed during the 5th EVT (after the completion of two weeks of endoscopic therapy) at latest; if AL had not improved endoscopically and radiologically by that point, no further EVT was offered and alternative therapies were considered.

Results

In total, 8 patients (5 male; mean age 61) underwent EVT over the above-mentioned period. Six of them underwent Ivor-Lewis esophagectomy with esophago-gastric anastomosis and 2 underwent total gastrectomy with esophago-jejunal anastomosis. Five of them had received neo-adjuvant chemotherapy prior to surgery. EVT was placed intracavitary initially if technically possible (5 patients) and then intraluminally as the leak site decreased in size, or directly intraluminally (3 patients) if the anastomotic leak was <5mm and could not accommodate our smallest in-house device.
EVT was successful in 7 out of 8 patients (success rate 87,5%). Only one patient had no response on day 15, and after discussion at our MDT he was offered a redo surgery (this was a patient with esophago-jejunostomy) with good final result. The mean time between AL diagnosis and first EVT was 9 days. A median of 6 endoscopic EVT procedures was needed (range 2- 15) for total AL closure. There were no technical issues related to the device itself, all patients tolerated EVT well and no adverse event directly related to EVT was recorded (the one patient who underwent reconstructive surgery experienced worsening of extraluminal cavity/collection but this was deemed secondary to the leak itself rather than EVT).

Conclusion

EVT is an effective therapy for UGI post-surgical anastomotic leaks, with a success rate of 87,5% in our experience. It is demanding in terms of resources (high number of repeat endoscopies needed), but it is well tolerated and safe.