A systematic review and meta-analysis of postoperative use of NSAIDs and risk of anastomotic leak

A systematic review and meta-analysis of postoperative use of NSAIDs and risk of anastomotic leak

Can J Surg 2020;63(4):E359-E361 | PDF

Arden M. Morris, MD, MPH; Larissa Temple, MD, MSc; Manoj Raval, MD, MSc; for the Evidence Based Reviews in Surgery Group

Key points about the article

Background: Enhanced recovery after surgery (ERAS) including use of nonsteroidal anti-inflammatory drugs (NSAIDs) is considered the standard of care for most abdominal operations. However, there is controversy around the association between these drugs and anastomotic dehiscence.

Study objective: To determine whether postoperative NSAID use was associated with intestinal anastomotic dehiscence.

Methods: The meta-analysis included published randomized controlled trials (RCTs) and observational studies that compared postoperative NSAID use with nonuse and reported on intestinal anastomotic dehiscence. Key databases searched were PubMed, EMBASE and CENTRAL, without date or language restriction.

Results: The study included 6 RCTs that included anastomotic dehiscence as a secondary outcome (n = 473) and 11 observational studies (n = 20 184). Among RCTs, pooled analyses revealed that NSAID use was clinically but not statistically associated with anastomotic dehiscence (relative risk [RR] 1.96, 95% confidence interval [CI] 0.74–5.16, I2 = 0%). Among observational studies, pooled analyses revealed that NSAID use was significantly associated with anastomotic dehiscence (odds ratio [OR] 1.46, 95% CI 1.14–1.86, I2 = 54%).

Conclusion: The RCTs included in the metaanalysis were of high methodologic quality, although none included a prespecified definition of anastomotic dehiscence. Five of the 11 observational studies did not include a clear definition of criteria for anastomotic dehiscence. None accounted for preoperative NSAID use. The authors concluded that postoperative nonselective NSAID use compared with nonuse was associated with significantly higher odds of intestinal anastomotic dehiscence. We surmise that this meta-analysis represents the best available evidence to date. However, the small negative impact of NSAID use on anastomotic leak must be weighed against the potential benefits in postoperative pain and recovery.


Affiliation: From the Department of Surgery, Stanford University School of Medicine, Stanford, CA (Morris); the Department of Surgery, St. Paul’s Hospital, Vancouver, BC (Raval); and the Department of Surgery, University of Rochester Medical Center, Rochester, NY (Temple).

Competing interests: None declared.

Contributors: All authors contributed substantially to the conception, writing and revision of this article and approved the final version for publication.

Members of the Evidence Based Reviews in Surgery Group: Chad G. Ball, Nancy N. Baxter, Mantaj Brar, Carl J. Brown, Prosanto K. Chaudhury, Indraneel Datta, Sandra de Montbrun, Justin Dimick, Elijah Dixon, G. William N. Fitzgerald, Samantha Hendren, Lillian S. Kao, Andrew Kirkpatrick, Steven Latosinsky, Robin S. McLeod, Arden M. Morris, Jason Park, Timothy M. Pawlik, Manoj Raval, Kjetil Soreide, Malin Sund, Larissa Temple, Bas Wijnhoven, Desmond Winter.

DOI: 10.1503/cjs.021419

Correspondence to:A.M. Morris, Stanford University School of Medicine, 875 Blake Wilbur Dr Clinic B MC 6560, Stanford CA 94305 ammsurg@stanford.edu