Background and purpose To investigate the relationship between the radiation dose to the anastomotic region and postoperative anastomotic leakage rates after McKeown oesophagectomy with cervical anastomosis. Materials and methods Between January 2017 and December 2019, 164 consecutive patients undergoing trimodal therapy including neoadjuvant chemoradiotherapy and sequential McKeown oesophagectomy were included. The demographic and clinical patient data were collected. Additionally, the radiation dose to the regions including the mediastinum, airway, gastric fundus and anastomotic region was recalculated. Results Twelve patients presented with anastomotic leakage, accounting for 7.3% (12/164) of the cohort. The anastomoses were located in the radiation field for all patients with anastomotic leakage (12/12, 100%), and for 61.8% (94/152) of those without (P = 0.009). Higher radiation doses, including the D50 and the mean, maximal and minimal doses to the oesophageal anastomotic region were found in the anastomotic leak group. Subgroup analysis between patients with end to end (ETE) anastomosis and ETE intussusception anastomosis revealed a lower anastomotic leakage rate in the latter after 1:1 ratio propensity score-matching (10.4% vs. 1.3%, P = 0.034). Conclusion Anastomosis location in the radiation field and a higher radiation dose to the oesophageal anastomotic region were associated with the occurrence of anastomotic leakage after trimodal therapy. Compared with ETE anastomosis, ETE intussusception anastomosis might reduce the occurrence of anastomotic leakage after neoadjuvant chemoradiation and subsequent McKeown oesophagectomy.