Risk of perioperative bleeding related to highly selective cyclooxygenase-2 inhibitors: A systematic review and meta-analysis

Semin Arthritis Rheum. 2017 Feb;46(4):520-528. doi: 10.1016/j.semarthrit.2016.07.008. Epub 2016 Jul 20.


Background: Several literatures reported that highly selective cycloxygenase-2 inhibitors (COX-2 inhibitors) had no effect on platelet function. However, some experts suggested stopping all non-steroidal anti-inflammatory drugs (NSAIDs), including COX-2 inhibitors at least five elimination half-lives prior to surgery. We, therefore, systematically summarized the risk of clinical bleeding or platelet dysfunction in healthy or surgical subjects who received COX-2 inhibitors.

Methods: Two electronic databases, MEDLINE and EMBASE, were searched for randomized, controlled studies published during 1980-December 2015. Additionally, manual search was performed to identify potential eligible studies. Intervention was perioperative use of any available COX-2 inhibitors in current practice (celecoxib, parecoxib, or etoricoxib), compared to non-selective NSAIDs, analgesics, or placebo. Two independent reviewers selected eligible studies, extracted the data, and assessed the quality of the included studies. The primary outcome was postoperative bleeding. All analyses were performed using RevMan-5.3.

Results: Of 3900 abstracts reviewed, 35 studies met the inclusion criteria. The data from 16 out of 35 studies were used in this meta-analysis, and outcomes of other 19 remaining studies were descriptively summarized. COX-2 inhibitors did not significantly increase the risk of postoperative bleeding events (relative risk or RR = 0.92; 95% confidence interval or CI: 0.63-1.33; p = 0.65), intraoperative blood loss (mL) (weighted mean difference or WMD = -4.38; 95% CI: -14.69 to 5.92; p = 0.4), postoperative blood loss (mL) (WMD = -13.89; 95% CI: -30.24 to 2.47; p = 0.10), and 24-h postoperative hemoglobin loss (g/dL) (WMD = 0.47; 95% CI: -0.14 to 1.09; p = 0.13). Platelet function analyzer closure time (second) significantly decreased with the use of COX-2 inhibitors (WMD = -22.22; 95% CI: -44.03 to -0.41; p < 0.00001). In the 19 remaining studies, COX-2 inhibitors did not significantly increase risk of bleeding in both clinical and laboratory outcomes.

Conclusion: Highly selective COX-2 inhibitors did not significantly increase the risk of intraoperative, postoperative bleeding, or blood loss. They also had no significant effect on platelet function. Therefore, perioperative, single dose, or short course of COX-2 inhibitors can be safely used in individuals who are undergoing surgery.

Keywords: Highly selective COX-2 inhibitor; Non-selective non-steroidal anti-inflammatory drug; Periooperative bleeding; Platelet function; Systematic review.

Publication types

  • Meta-Analysis
  • Review
  • Systematic Review

MeSH terms

  • Blood Loss, Surgical / statistics & numerical data*
  • Celecoxib / therapeutic use
  • Cyclooxygenase 2 Inhibitors / therapeutic use*
  • Etoricoxib
  • Humans
  • Isoxazoles / therapeutic use
  • Postoperative Hemorrhage / epidemiology*
  • Pyridines / therapeutic use
  • Risk Factors
  • Sulfones / therapeutic use


  • Cyclooxygenase 2 Inhibitors
  • Isoxazoles
  • Pyridines
  • Sulfones
  • parecoxib
  • Celecoxib
  • Etoricoxib