NSAIDs Use and Postoperative Medical Complications
Oral or IV NSAIDs administered preoperatively, intraoperatively, or postoperatively do not appear to increase the risk of medical complications following primary TJA; however, providers should consider patient comorbidities, the type of NSAID administered, dose, and duration of administration.
Anesthesia and Analgesia in Total Joint Arthroplasty (2020)
Developed by: American Association of Hip and Knee Surgeons, American Society of Regional Anesthesia and Pain Medicine, American Academy of Orthopaedic Surgeons, The Hip Society, and The Knee Society

Rationale

Among the reviewed high and moderate quality randomized clinical trials comparing perioperative oral NSAIDs and placebo, twelve studies reported on medical complications related to the administration of NSAIDs.[2-5, 7-9, 11-15] Qualitative examination demonstrated no consistent difference between oral selective COX-2 (includes selective [i.e. Celecoxib] and preferential [i.e. Meloxicam] COX-2 inhibitory agents) NSAIDs, oral non-selective (COX-1 and -2 inhibitory agents) NSAIDs, and placebo with the exception of a lower incidence of postoperative fever with patients receiving an oral NSAID. Direct meta-analysis was capable of being performed comparing various complications between perioperative NSAIDs and placebo, which showed no significant difference with regards to any adverse event (0.93 relative risk; 95% confidence interval of 0.85 to 1.02), vomiting (0.82 relative risk; 95% confidence interval of 0.52 to 1.31), nausea (0.84 relative risk; 95% confidence interval of 0.68 to 1.04), blood loss (-0.23 standard mean difference; 95% confidence interval of -0.54 to 0.08), pruritus (1.73 relative risk; 95% confidence interval of 0.96 to 3.13), urinary retention (1.24 relative risk; 95% confidence interval of 0.34 to 4.59), and sedation (0.46 relative risk; 95% confidence interval of 0.16 to 1.26). Similar to oral NSAIDs, direct meta-analysis of medical complications between IV ketorolac and placebo were not significant with regards to any adverse events (0.94 relative risk; 95% confidence interval of 0.59 to 1.50), nausea (0.89 relative risk; 95% confidence interval of 0.70 to 1.12), vomiting (0.73 relative risk; 95% confidence interval of 0.47 to 1.14), blood loss (-0.14 standard mean difference; 95% confidence interval of -0.46 to 0.17), pruritus (0.50 relative risk; 95% confidence interval of 0.22 to 1.12), urinary retention (0.75 relative risk; 95% confidence interval of 0.43 to 1.32), or respiratory depression (-0.05 standard mean difference; 95% confidence interval of -0.28 to 0.18).

           Despite the evidence favoring oral and IV NSAIDs in the qualitative and quantitative analysis of numerous high and moderate quality studies to reduce postoperative pain and opioid consumption, the gastrointestinal and renal safety profile of oral and IV NSAIDs have not been thoroughly studied in patients following primary TJA. Although nausea and vomiting were frequently reported among the studies, more severe complications including upper gastrointestinal bleeding and acute renal failure were not reported. It is possible the lack of reporting an upper gastrointestinal bleed is due to the rarity of the complication. As a result, clinicians should consider the safety of perioperative NSAIDs as it relates to severe gastrointestinal and renal failure complications. Therefore, the work group downgraded the recommendation strength by only assigning a limited strength to the recommendation.