Non-steroidal anti-inflammatory drugs in colorectal surgery: A risk factor for anastomotic complications?

In a recent article, Gorissen et al report on 795 patients with primary colorectal anastomosis operated on during the period 2008-2010 for different colorectal conditions at two centres. The leakage rate was significantly higher among patients who were administered non-steroidal anti-inflammatory drugs (NSAIDs) in the perioperative course. A dose-response relationship could also be traced, where longer NSAID use yielded a higher risk of anastomotic breakdown. However, as this study is observational in design, confounding by indication may be present and there is also a risk of residual confounding from unmeasured covariates. Moreover, the question whether different affinity for the cyclooxygenase enzyme is important in different NSAIDs seems to be largely unanswered. The results, conclusions and clinical relevance of the aforementioned study, including the possible effects of different types of NSAIDs, are discussed. While acknowledging that this study represents the best attempt so far in establishing the causal relationship between perioperative NSAID use and anastomotic leakage, the need for further research in this important area is underlined.