OGC P37 Towards an era of eliminating leak related morbidity from major oesophago-gastric surgery

Abstract

Abstract Background A leak following oesophago-gastric surgery remains a life-threatening complication associated with a long intensive care stay and high mortality and morbidity. We hypothesised that Endoluminal Vacuum Therapy (EVT) could be used prophylactically to provide source control, prevent mediastinitis and minimise sepsis related morbidity if breakdown of an oesophageal or gastric anastomosis or repair occurred. In the first instance, this could be a particularly attractive adjunct in cases where technical and/or patient factors deem an anastomosis or repair as high risk. Methods We collected data prospectively from all patients where a prophylactic EVT device was placed during elective oesophago-gastric surgery. Primary outcome measures included length of stay (LOS) and post-operative leak rate. Prophylactic EVT was provided using an ad-hoc EVT device constructed by suturing a piece of V.A.C granufoam (KCI) to the end of an 18Fr NG tube and positioning this intraluminally across the anastomosis or area of repair at the time of surgery using endoscopic guidance. Continuous negative pressure of between 75 and 125mmHg was applied. Following removal of the EVT device, assessment of the integrity of the anastomosis or repair was made using one, or a combination of endoscopic assessment, CT with oral contrast, and clinical assessment. Results Prophylactic EVT was utilised in 9 patients; M:F ratio=7:2, median age 72 years (range 50–80). 8 patients had ASA 3, and 1 had ASA 2. The index surgery was resection of an epithelial cancer in 7 patients (3 partial/total gastrectomies, 4 oesophagectomies), resection of a gastric GIST in 1 patient, and a Hellers myotomy in 1 patient. All anastomosis/repairs were deemed high risk due to technical(n=3), patient(n=4) or combined (n=2) factors. Prophylactic EVT was placed for a median of 6 days (range 4–10). The integrity of the anastomosis/repair was assessed using both CT and gastroscopy (n=3), CT alone (n=3), gastroscopy alone (n=2), and clinical grounds (n=1). One patient was found to have a full thickness oesophageal defect at their first endoscopic assessment. This was managed with a further period of intraluminal EVT until the oesophageal defect healed. The patient remained clinically well throughout and was managed entirely with ward-based care. The median LOS was 11 days (range 7–96). The two longest stays were secondary to a prolonged chyle leak and longer term sequelae of a life threatening on-table bleed. In all cases the site of repair remained healthy with no evidence of early stricturing. Conclusions Prophylactic EVT therapy was safe and did not result in any short-term adverse events in this small cohort of patients. Although it may not prevent breakdown at the site of oesophageal anastomosis/repair and is not an alternative to sound surgical technique, the use of prophylactic EVT in high risk anastomosis could be beneficial in preventing life threatening sepsis and associated systemic effects resulting from inadequate source control. Current studies are under way to look at longer term outcomes in the form of prospective cohort studies and randomised controlled trials. This provides the oesophago-gastric community the unique opportunity to collectively eliminate life threatening leak related morbidity and mortality for once and for all.