This site is intended for health professionals only

Cox-2 inhibitors for postoperative pain


Chad T Whelan
Assistant Professor of Medicine
Department of Medicine
University of Chicago

Postoperative pain is common and, unfortunately, often not adequately controlled. Poorly controlled pain results in lower patient satisfaction as well as poor surgical outcomes.(1,2) Narcotics, which are the primary agents used, may not adequately control postoperative pain and frequently have side-effects. Given the prevalence of postoperative pain and our present inadequacies in managing it, improved methods for controlling postoperative pain are needed. Non-narcotic pain-modifying agents may play an important adjuvant role in postoperative pain control. Currently, the most promising medications for this role are the cyclo-oxygenase (cox)-2 inhibitors, which have analgesic and anti-inflammatory effects, as well as excellent side-effect profiles.

Postoperative pain can often be controlled with narcotic analgesics. Establishing postoperative narcotic-based pain management pathways has been shown to decrease pain, improve satisfaction with pain control and even aid surgical recovery.(3) However, even with well-designed clinical pathways, postoperative pain is not always controlled with narcotics alone. Additionally, the side-effects of narcotics, such as nausea, vomiting, constipation, sedation and confusion, are common. Nausea and vomiting can occur in 25–50% of patients requiring narcotics postoperatively.(1,4)
Treatment and prevention of these side-effects often requires additional medications, all of which have their own side-effects. Recovery from surgery can be prolonged by inadequate pain control or by the adverse effects of narcotics. This prolonged recovery often results in extended length of stays, and may lead to additional complications.(1,5) Finally, increased early postoperative pain is a strong predictor of extended postoperative pain.(6) Narcotics have an important role in the management of postoperative pain. However, their use is complex, often requiring multiple modes of delivery, including patient-controlled analgesia (PCA) pumps, and still leave significant room for better pain control in this high-risk pain population.

Given these limitations to postoperative narcotic use, improved methods of managing pain are needed. While most efforts to date have focused on treating pain after it has started, it may be more effective to prevent some of the components of pain. While the neural mechanisms of postoperative pain are complex, our understanding of the pathways involved is increasing. The trauma of surgery induces cox-2, leading to prostaglandin synthesis, which has been shown to be a contributor to postoperative pain.  Thus, pharmacological inhibition of this pathway could not only be analgesic, but it may also reduce the induction of pain. Preoperative NSAIDs have been shown to be effective at blunting prostaglandin synthesis and improving postoperative pain control while reducing narcotic use.(7) However, NSAIDs are rarely used during the perioperative period, due to the increased bleeding risk with NSAIDs (a doubling of blood loss in elective hip arthroplasty(8)). In fact, patients are usually told to stop taking any NSAIDs for a week before surgery.

Cox-2 inhibitors theoretically maintain the advantages of NSAIDs without the risk of increased bleeding, with evidence emerging from several recent clinical trials. In randomised double-blind placebo controlled trials involving multiple types of surgeries, including orthopaedic and open lower abdominal surgery, cox-2 inhibitors were effective at improving pain control, reducing narcotic use and reducing narcotic side-effects when given preoperatively or pre- and postoperatively.(4,9,10) In selected studies, cox-2 inhibitors also improved postoperative pulmonary function, better postoperative range of motion after orthopaedic surgery and shortened the required duration of physical therapy. In a study by Buvanendran et al, published in JAMA, 70 patients were randomly assigned to rofecoxib or placebo given both pre- and postoperatively for a total knee replacement.(4) The treatment protocol called for 50mg of rofecoxib 24h before surgery and a second dose 1–2h before the operation. After surgery, 50mg of rofecoxib was given daily for 5 days, followed by 25mg for an additional 8 days. Rofecoxib-treated patients had clinically impressive benefits in pain scores, gastrointestinal side-effects and better range of motion, even when using fewer overall narcotics (see Table 1). The treatment arm was also significantly more satisfied with their pain control in the hospital. This improved satisfaction persisted at a month postoperatively.(4) Although cox-2 inhibitors have known complications of nephrotoxicity, gastrointestinal upset and possible cardiovascular complications, significant increases in these adverse events did not occur in any of the trials.


While the published clinical trials appear to be very promising, there are some limitations. A large majority, but not every trial, demonstrated clinical benefit. Importantly, these trials were small, with treatment arm groups of 30–70 patients.(11) Given these sample sizes, they were not powered to detect differences in uncommon side-effects. While adverse effects are rare with cox-2 inhibitors, there are some potentially very serious effects to consider. Therefore, for patients at risk for these potentially serious adverse effects, particularly renal or cardiovascular complications, care should be exercised when using cox-2 inhibitors. The optimal perioperative dosing strategies for cox-2 use are also unknown. In the published studies, dosing ranged from a single preoperative dose to several preoperative doses and days of postoperative use. Further studies should help define more clearly the risks and optimal dosing of perioperative cox-2 inhibitor use.

Given the recent publication of many clinical trials demonstrating the effectiveness of cox-2 inhibitors for the treatment of postoperative pain, the perioperative use of these agents will likely increase. While evidence is mounting to support their use, the data are not complete. Pharmacists should play an active role in assuring that the use of these agents for postoperative pain is appropriate. Clinical pathways for perioperative pain management should be utilised in which cox-2 inhibitors could play a role at least preoperatively and perhaps in the postoperative period for some patients. Specific guidelines identifying appropriate patients and procedures should be created to minimise the potential serious renal and cardiovascular side-effects of these medications. Pharmacists and their institutions should consider measuring the effects of cox-2 use in their own institution during the perioperative period, with special attention to patient satisfaction, resource utilisation/ cost-effectiveness and side-effects.


  1. Acute Pain Management: Operative or Medical Procedures and Trauma Clinical Practice Guideline No. 1. AHCPR Publication No. 92-0032: February 1992.
  2. Ballantyne JC, et al. Anesth Analg 1998;86:598-612.
  3. Brodner G, et al. Euro J Anaesthesiol 2000;17:566-75.
  4. Buvanendran A, et al. JAMA 2003;290:2411-8.
  5. Capdevila X, et al. Anesthesiology 1999;91:8-15.
  6. Perkins FM, Kehlet H. Anesthesiology 2000;29:1501-10.
  7. Reasbeck PC, et al. Lancet 1982;2:15-8.
  8. Robinson CM, et al. J Arthroplasty 1993;8:607-10.
  9. Sinatra RS, et al. Anesth Analg 2004;98:135-40.
  10. Malan TP, et al. Anesthesiology 2003;98:950-6.
  11. Gilron I, et al. Anesthesiology 2003;99: 1198-1208.

American Society of Anesthesiologists
American Pain Society

Latest Issue

Be in the know
Subscribe to Hospital Pharmacy Europe newsletter and magazine