Effect of Perioperative Dexamethasone and Different NSAIDs on Anastomotic Leak Risk: A Propensity Score Analysis.

Abstract

Perioperative use of nonsteroidal anti-inflammatory drugs (NSAIDs) is associated with risk of anastomotic leak (AL). However, concomitant use of other drugs could infer a bias in risk assessment. Thus, we aimed to interrogate the risk of AL associated with NSAIDs and steroids used perioperatively. This study includes a consecutive series of patients having surgery involving an intestinal anastomosis from Jan 2007 to Dec 2009. Data records included demographic, perioperative, and surgical characteristics; AL rates; and use of NSAIDs and steroids. Risk of leak were estimated using unadjusted and multivariable (propensity score)-adjusted logistic regression models and reported as odds ratios (ORs). A total of 376 patients underwent 428 operations of which 67 (15.7 %) had AL. With no medication receivers as reference, the OR for leak when adjusted for age, sex, and propensity score was 1.07 (p = 0.92) for ketorolac, 1.63 (p = 0.31) for diclofenac and 0.41 (p = 0.19) for dexamethasone. Risk was increased for malignancy (OR 1.88, p = 0.023), use of a vasopressor (OR 2.52, p = 0.007), blood transfusions (OR 1.93, p = 0.026), and regular use of steroids (OR 7.57, p = 0.009). Other factors than perioperative drugs are crucial for risk of AL. Perioperative dexamethasone was associated with a nonsignificant reduced risk of AL.