Drug Insight: Cyclo-Oxygenase-2 Inhibitors--A Critical Appraisal

Burkhard Hinz; Bertold Renner; Kay Brune


Nat Clin Pract Rheumatol. 2007;3(10):552-560. 

In This Article


Since the hasty voluntary withdrawal of rofecoxib, discussions about COX2 inhibitors have often been characterized more by emotions than scientific evidence. These agents cause significantly fewer severe side effects in the gastrointestinal tract than NSAIDs, and recent studies indicate that COX2 inhibition elicits significantly less macroscopic injury to the small intestine than a NSAID plus a proton-pump inhibitor. COX2 inhibitors should, therefore, be preferentially used by patients with gastrointestinal risk factors (ulcer history, concomitant medication with oral anticoagulatives or glucocorticoids). Other advantages of these drugs include the lack of bleeding that they cause, and the fact that they can be tolerated by patients with aspirin-induced asthma.

Well-designed, placebo-controlled trials assessing the off-label long-term use of COX2 inhibitors in patients with a history of colorectal adenomas led to the observation that these drugs are associated with an increased rate of cardiovascular adverse effects. The MEDAL study, however, showed that NSAIDs (diclofenac) and COX2 inhibitors (etoricoxib) result in a similar cardiovascular risk, suggesting that there is presently no rationale for a further differentiation of these groups of drugs in terms of cardiovascular toxicity. Further randomized controlled trials are needed to confirm if this finding is true for other COX2 inhibitors. Moreover, nonacidic antipyretic analgesics, such as acetaminophen and dipyrone, which have been claimed to exhibit no significant action on peripheral prostaglandin formation, elicit a substantial inhibition of COX2 in vivo[60,61] and should, therefore, also be thoroughly analyzed for cardiovascular side effects. Remarkably, an analysis of the Nurses' Health Studies showed that acetaminophen (at daily doses >500 mg) was associated with a significantly higher risk of developing hypertension in comparison with no use.[62] The relative risk of hypertension associated with acetaminophen was, in fact, similar to that of NSAIDs. In accordance with this, a large, prospective study showed that use of more than 15 tablets of acetaminophen per week confers nearly the same risk as NSAIDs for cardiovascular events.[63]

Present research is focusing on finding the lowest dose of COX2 inhibitors and the shortest duration for their use sufficient for analgesia. In addition, work is ongoing to identify and characterize laboratory markers (e.g., the aminoterminal part of B-type natriuretic peptide [NT-pro BNP]) that could be used to identify patients at cardiovascular risk before a decision is taken concerning the use of COX2 inhibitors.

In summary, both COX2 inhibitors and NSAIDs are only conditionally appropriate for long-term treatment of musculoskeletal pain, but—in consideration of newly defined contraindications and warnings—still remain important tools in pain therapy.[40,64] The not easily addressable goal of medical practice is to find out the least risky therapy for the individual patient.

Note added in proof. While this Review was in press, Australia's drugs regulator—the Therapeutic Goods Administration (TGA)—cancelled the registration of lumiracoxib (100 mg, 200 mg, and 400 mg tablets) because of concerns over hepatic toxicity (eight cases of serious liver adverse effects, including two deaths, with another two patients needing liver transplants) in patients receiving the drug at daily doses of ≥200 mg.[66] Approximately 60,000 patients have used lumiracoxib in Australia, and most of these have been prescribed 200 mg tablets for the management of osteoarthritis (OA). According to the manufacturer, the 100 mg dose of lumiracoxib, which is the recommended dose worldwide for the treatment of OA, has not been associated with an unexpected incidence of liver-related adverse effects in an OA population; however, a close analysis will be necessary to confirm this statement. Moreover, given that the peak plasma concentration of 200 mg lumiracoxib[67] is approximately 100-fold higher than the concentration expected to cause half-maximal cyclooxygenase-2 inhibition (i.e. 0.13 μmol/l) in the human whole blood assay,[68] doses <100 mg should also be considered for pain therapy.

CLICK HERE for subscription information about this journal.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.