Nasogastric tube after oesophagectomy and risk of anastomotic leak: a Nordic, multicentre, open-label, randomised, controlled, non-inferiority trial

Background: Oesophagectomy, a corner stone in curative treatment of oesophageal cancer, is a complex procedure with high complication rates. Postoperative gastric tube decompression is debated and some centres are abandoning routine nasogastric (NG) tube use. We hypothesised that postoperative NG tube removal is non-inferior to five days of NG tube decompression, with regard to the risk of anastomotic leak. Methods: In this open-label, non-inferiority randomised controlled trial across 12 hospitals in Sweden, Norway, Denmark and Finland, participants treated for oesophageal or gastroesophageal junctional cancer with oesophagectomy were randomly assigned (1:1) to no postoperative NG tube or five days of NG tube decompression. Anastomotic leak was the primary outcome and secondary outcomes included pneumonia and length of hospital stay. Analyses were performed on the intention to treat and per protocol populations and non-inferiority for anastomotic leak was defined as a risk difference below 9%. ISRCTN.com registration ISRCTN39935085. Findings: Between January 1st 2022 and March 27th 2024, 448 patients were randomly assigned, 217 to no postoperative NG tube and 231 to five days NG tube treatment. The mean age was 67.5 (standard deviation (SD) 9.8) years and 367 (81.9%) were males. Non-inferiority with regard to anastomotic leak for no NG tube decompression could not be shown with 48 patients (22.1% (95% confidence interval (CI) 16.8%, 28.2%)) having anastomotic leak compared to 35 (15.2% (95% CI 10.8%, 20.4%)) with five days of NG tube decompression, a risk difference of −7.0% (95% CI −14.4%, 0.00%), p_{non-inferiority} 0.30. In a Supplementary analysis, patients had a lower risk of anastomotic leak if postoperative NG decompression was used. Rate of other complications, e.g., pneumonia, were similar between groups. In a per-protocol analysis, the risk difference was −11.3% to the advantage of NG tube (95% CI, −19.1, −0.3%). Interpretation: We could not establish safety (increased risk of anastomotic leak) and therefore do not support omission of NG tube after oesophagectomy. Funding: This trial was funded by the Swedish Cancer Society and the Nordic Cancer Union.