Exposure to oestrogen and risk of anastomotic leakage after colorectal cancer surgery - A clue to the different leak rates in men and women.

BACKGROUND: Colorectal anastomotic leakage is consistently more common in men, regardless of tumour location. This fact is largely unexplained but might be a consequence of biological differences including hormonal exposure and not only related to anatomy. METHODS: This was a retrospective, nationwide registry-based observational study of post-menopausal women operated for colorectal cancer with an anastomosis between 2007 and 2016. Hormonal exposure before surgery, as defined by prescribed drugs affecting oestrogen levels, was related to postoperative anastomotic leakage, using mixed-effects logistic regression models with adjustment for confounding. Odds ratios (ORs) with corresponding 95% confidence intervals (CIs) were derived. In addition, separate estimates according to tumour location were computed, and a sensitivity analysis excluding topical oestrogen hormone exposure was conducted. RESULTS: Some 16,535 post-menopausal women were included, of which 16.2% were exposed to drugs increasing oestrogen levels before surgery. In this exposed group compared to the unexposed, leak rates were 3.1 and 3.8%, respectively. After adjustment, a reduction of anastomotic leakage in the exposed group was detected (OR: 0.77; 95% CI: 0.59-0.99). This finding was largely attributed to the rectal cancer subgroup (OR: 0.55; 95% CI: 0.36-0.85), while the exclusion of topical oestrogen drugs further reduced the estimates of the main analysis (OR: 0.63; 95% CI: 0.38-1.02). CONCLUSIONS: Anastomotic leakage rates are lower in women exposed to hormone replacement therapy before surgery for colorectal cancer, which might explain some of the difference in leak rates between men and women, especially regarding rectal cancer.