To the Editor I would like to make several comments in relation to the article by Hakkarainen et al1 recently published in JAMA Surgery. Being part of a hospital trust that performs large numbers of colorectal surgical procedures, we have a specific interest in the outcomes of this study.1 The benefits of nonsteroidal anti-inflammatory drugs (NSAIDs) with regard to the opioid-sparing effect cannot be denied. However, some surgeons are concerned about the risk of anastomotic leaks; therefore, they prefer not to use NSAIDs, despite the evidence of leakage being weak and the benefits of NSAID use being very clear. The authors1 acknowledge several limitations, but these do not diminish the use of NSAIDs in any way. When researchers are examining an outcome due to a particular drug, it is of paramount importance that the type of drug (ie, the name of the NSAID), the subtype (cyclooxygenase 1 vs cyclooxygenase 2), the dose, and the duration of therapy are known. Surely, at a clinical level, an anastomotic leak 85 days after surgery cannot be blamed on 2 to 3 doses of NSAIDs given postoperatively? Without any further breakdown of the data on the NSAID therapy that was used, no actual clinical link between the 2 events can be determined.