Rheumatology Advance Access originally published online on February 19, 2007
Rheumatology 2007 46(5):730; doi:10.1093/rheumatology/kel429
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EDITORIALS |
Do case control studies on coxibs tell us anything new?
Correspondence to: Dr Brune. E-mail: Kay.Brune{at}pharmakologie.med.uni-erlangen.de
If one reads the voluminous manuscript by Rahme and Nedjar (this issue) one gets—at the first glance—the impression that this is all known:
- Paracetamol and celecoxib exert the least damaging effect on the GI-tract and the cardiovascular system.
- Neither compound goes along with an increased cardiovascular risk in patients not receiving low dose aspirin.
- Low dose aspirin increases the GI-toxicity of all compounds with the exception of naproxen.
- Cardiovascular events occur more frequently in the groups treated with low dose aspirin.
- The incidence of cardiovascular events is particularly high if given together with ibuprofen and particularly low in combination with naproxen.
But then, after looking at all the data presented in the context of other recent findings, interesting new insights arise:
- The only drug going along with significantly more GI-bleeds is naproxen. Celecoxib is as harmless as acetaminophen. Diclofenac appears to cause fewer GI-toxicity than rofecoxib. Checking the data cautiously one observes, however, that almost double as many patients were under gastroprotective therapy when receiving traditional, non-steroidals as compared with those under coxib treatment. Consequently we may still assume that coxibs are better tolerated by the GI-tract.
- Altogether there are very few serious cardiovascular events. Interestingly, the hazard ratio is higher if aspirin is given together with cyclooxygenase inhibitors—particularly when added to ibuprofen treatment—but not when naproxen was administered. Again, an interesting observation that indicates that the patient population with higher cardiovascular risks will more often receive aspirin and, despite this treatment, encounter more cardiovascular events. Likewise patients with known GI-risks may have received coxibs preferentially and consequently will produce more unwanted drug effects (UDE) (rofecoxib) despite COX-2 selective action. The increase in GI-toxicity by aspirin is not seen when given with naproxen, which may be explained by the fact that naproxen is the only drug aside of low dose aspirin that blocks COX-1-dependent platelet aggregation sufficiently long enough to exert measurable anti-thromboembolic activity. This contention agrees well with the fact that ASS does not increase the GI-toxicity of naproxen in contrast to all other selective and non-selective inhibitors of prostaglandin production.
- This brings us to the third interesting conclusion: low dose aspirin does increase the propensity for GI-bleeds with almost all (except naproxen) cyclooxygenase inhibitors. The putative advantage of selective inhibitors with respect to GI-bleedings (particularly by celecoxib) vanishes completely when low dose aspirin is given. To what extent aspirin may protect from cardiovascular events is not obvious from this investigation. The difference between users of aspirin and non-users may well be explained by the higher risk in the former group.
Finally the data indicate that it is unlikely that combining blockade of cyclooxygenase-1 in platelets by aspirin or high doses of naproxen (for cardiac protection) plus inhibition of H+-secretion by proton pump inhibitors will solve all our problems. Naproxen shows the highest incidence of GI-toxicity despite the use of PPIs. The good news is that in patients receiving naproxen, the additional administration of aspirin appears superfluous, but not dangerous. Quite in contrast, aspirin plus ibuprofen increases cardiovascular toxicity substantially—which may be taken as additional proof for the well-known, direct interference of ibuprofen with aspirin activity at cyclooxygenase-1 in platelets.
K.B. has had consultancies with Bayer, Merck, Novartis, Pfizer and Sanofi.
|
Accepted 30 November 2006
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||