Abstract
BACKGROUND: Upper gastrointestinal tract anastomotic leaks are associated with increased morbidity and mortality, especially in the 30-day post-operative period. Techniques to reduce tension at the anastomotic site have been postulated to reduce leak rates. Our study investigates if continuous nasojejunal suction (via Freka trelumina tube) compared with standard of care passive free drainage via nasogastric tube results in reduced anastomotic leaks. METHODS: A retrospective cohort study was performed on all patients undergoing upper gastrointestinal tract resections with primary high-risk anastomosis (proximal partial, total/extended total gastrectomies and oesophagectomies) over the period between 2012 and 2023 between two major tertiary centres for malignancy. Comparative analysis of outcomes was performed between patients who did or did not receive prophylactic continuous suction, with the primary outcome being the incidence of anastomotic leak. Secondary outcome was the association between patient baseline characteristics and past medical history with anastomotic leaks. Statistical analysis was performed using SPSS v25.0. RESULTS: Over the 10-year period there were 351 operative cases, with prophylactic suction applied to 62.3% (n = 219) of patients. Overall incidence of anastomotic leaks was 10.3%, with higher proportions from 2-stage oesophagectomies (12.2%), compared with 3-stage oesophagectomies (5.9%) and gastrectomies (6.8%). In the overall cohort, incidence of anastomotic leaks was lower in the suction group (8.7% vs. 12.9%); however, this was not statistically significant (p = 0.210). Further subgroup analyses of 2-stage oesophagectomies showed marginal significance (9.0% vs. 18.9%, p = 0.050). Statistically significant risk factors for anastomotic leaks include any smoking history, elevated BMI, history of ischaemic heart disease, and prolonged operation duration. CONCLUSIONS: Continuous gastric conduit suction may reduce the incidence of anastomotic leaks in two-stage oesophagectomies; however, a randomized controlled trial is required to investigate this further.