Abstract

It is unclear whether gastrectomy or oesophagectomy offer better outcomes for gastro-oesophageal junction (GOJ) cancer. A total of 240 patients undergoing total gastrectomy (n = 85) or oesophagectomy (n = 155) for Siewert II-III GOJ adenocarcinoma were identified from a Swedish prospective population-based nationwide cohort. The surgical approaches were compared in relation to non-radical resection margins (main outcome) using multivariable logistic regression, providing odds ratios (ORs) and 95% confidence intervals (CIs), mean number of removed lymph nodes with standard deviation (SD) using ANCOVA, assessing mean differences and 95% CIs, and 5-year mortality using Cox regression estimating hazard ratios (HRs) and 95% CIs. The models were adjusted for age, sex, comorbidity, tumour stage, and surgeon volume. The non-radical resection rate was 15% for gastrectomy and 14% for oesophagectomy, and the adjusted OR was 1.61 (95% CI 0.68–3.83). The mean number of lymph nodes removed was 14.2 (SD ± 9.6) for gastrectomy and 14.2 (SD ± 10.4) for oesophagectomy, with adjusted mean difference of 2.4 (95% CI-0.2–5.0). The 5-year mortality was 76% following gastrectomy and 75% following oesophagectomy, with adjusted HR = 1.07 (95% CI 0.78–1.47). Gastrectomy and oesophagectomy for Siewert II or III GOJ cancer seem comparable regarding tumour-free resection margins, lymph nodes removal, and 5-year survival.