Abstract Aim The optimal operation for Siewert type II cancers is unclear, with both oesophagectomy and total gastrectomy being undertaken. This study aimed to investigate the overall survival and associated operative outcomes of both approaches. Method A retrospective cohort study was performed of all Siewert type II Gastro-Oesophageal Junctional (GOJ) resections from a single oesophagogastric centralised service between 2010 and 2017. The primary outcome was 5-year overall survival (OS). Secondary outcomes were recurrence rate and operative morbidity. Results A total of 163 patients were included; 127 (77.9%) oesophagectomy vs. 36 (22.1%) gastrectomy. The 5-year OS for oesophagectomy vs. gastrectomy were 50.4% vs 41.7% with a median survival of 57 months vs. 31 months respectively (p = 0.062). No differences in the rate of recurrence (44.9% vs. 33.3%, p = 0.216) or overall rate of complications (63.8% vs. 63.9%, p = 0.990) were identified. Yet, significantly more major (Clavien-Dindo III+) complications occurred in the gastrectomy compared to the oesophagectomy cohort (38.9% vs. 20.5%, p = 0.012), driven partially by a higher anastomotic leak rate (19.4% vs 9.4%, p = 0.099). Cox regression revealed no significant difference in the 5-year OS between the two groups when accounting for confounders (HR 0.906, 95% CI 0.339-2.060, p = 0.814). Only age (HR 0.971, 95% CI 0.944-0.999, p = 0.046) and pT stage (HR 1.794, 95% CI 1.097-2.935, p = 0.020) were determinants of 5-year OS. Conclusions When accounting for confounders there were no significant differences in the overall survival based on the two approaches. Further multi-centre prospective studies would be advantageous to establish a consensus on the best approach.

Surgeon‐level operative mortality is widely seen as a measure of quality after gastric and oesophageal resection. This study aimed to evaluate this alongside a compound‐level outcome analysis.