Long-term significance of an anastomotic leak in patients undergoing an ultra-low anterior resection for rectal cancer.

Background: Australia has one of the highest rates of colorectal cancer worldwide. Despite technological advances in colorectal surgery, anastomotic leaks (ALs) continue to cause significant morbidity and mortality. Ultra-low anterior resections (ULARs) carry the highest prevalence of AL. The aim of the study is to evaluate the incidence, treatment and consequences of AL following ULAR for colorectal cancer from a single colorectal unit.

Methods: This is a retrospective evaluation of prospectively collected data on patients undergoing ULAR following rectal cancer. The main end points include the prevalence and management of AL following initial operation and the morbidity, re-operation and mortality rates associated with AL. A stepwise logistic regression analysis and a multivariate analysis were performed to identify independent risk factors.

Results: A total of 467 patients underwent an ULAR. There were 32 (6.8%) ALs. Average follow-up time was 79 months. There were five subclinical leaks and only one (20%) required intervention. The overall survival rate at 5 years was 80% (95% confidence interval 58-91). On univariate analysis male sex was a risk factor for AL (P = 0.03). On multivariate analysis patients who had a complete response to radiotherapy were more likely to have a leak than the patients who had no radiotherapy (grade 4, odds ratio 4.0, 95% confidence interval 1.4-10.9, P = 0.01).

Conclusion: This study has highlighted the relevance of subclinical leaks and their associated morbidity. It identified that radiotherapy a risk factor for AL, but the response to radiotherapy is an even better predictor of leakage.