Background: It is uncertain which type of anastomosis carries the lowest risk of anastomotic insufficiency after oesophagectomy for oesophageal cancer. We aimed to compare handsewn with stapled anastomosis (any type, linear or circular), and handsewn end-to-side with handsewn end-to-end anastomosis.

Methods: This bi-national population-based cohort study included almost all patients (>95 %) who underwent oesophagectomy for cancer in Sweden from 2011 to 2020 or in Finland from 2004 to 2016. Multivariable logistic regression produced odds ratios (OR) with 95 % confidence intervals (CI), adjusted for age, sex, comorbidity, tumour histology, neoadjuvant chemo(radio)therapy, surgical approach, anastomosis location, hospital volume, and pathological tumour stage.

Results: Among 2166 study patients, 327 (15 %) had anastomotic insufficiency. The risk of anastomotic insufficiency was borderline significantly decreased in handsewn anastomosis compared to stapled anastomosis (OR = 0.79, 95 % CI 0.60-1.05). In patients who underwent minimally invasive oesophagectomy, handsewn anastomosis was associated with a decreased risk compared to stapled anastomosis (OR = 0.55, 95 % CI 0.35-0.85; n = 999), while no such association was found after open oesophagectomy (OR = 1.04, 95 % CI 0.72-1.51; n = 1167). There were no statistically significant associations with anastomotic insufficiency when comparing linear stapled with circular stapled anastomosis (OR = 1.27, 95 % CI 0.70-2.28; n = 736) or handsewn with circular stapled anastomosis (OR = 0.94, 95 % CI 0.63-1.40; n = 1324). Handsewn end-to-side anastomosis was associated with a borderline increased risk of anastomotic insufficiency compared to handsewn end-to-end anastomosis (OR = 1.61, 95 % CI 0.93-2.78; n = 786).

Conclusions: Regarding anastomotic insufficiency, handsewn anastomosis may be favourable compared to stapled in minimally invasive oesophagectomy for oesophageal cancer, while no such benefit was found for open oesophagectomy.