Exacerbation of inflammatory bowel disease by nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 inhibitors: Fact or fiction?

Abstract

The existence of a possible link between inflammatory bowel disease (IBD) and nonsteroidal anti-inflammatory drugs (NSAIDs) has been repeatedly suggested. Recently, a few studies have addressed the issue of a possible, similar effect by selective cyclooxygenase-2 inhibitors (COXIBs). The present article reviews the available scientific evidence for this controversial subject.

Key Words: COX-2 inhibitor; Inflammatory bowel disease; Non-steroidal anti-inflammatory drugs

TEXT

The use of nonsteroidal anti-inflammatory drugs (NSAIDs) has been associated with the onset of inflammatory bowel disease (IBD) or with a clinical flare-up of IBD in a number of case reports[1]. Both rectal and colonic frank ulcerations[2], small bowel ulcers[3] and intestinal, diaphragm-like strictures[1,3,4] have been reported after prolonged NSAID intake. On the other hand, no relationship is reported between NSAID treatment and exacerbation of underlying IBD[5,6].

The absence of controlled, prospective trials makes it difficult to draw definitive conclusions. Uncontrolled clinical experience suggests that anti-inflammatory agents can occasionally elicit relapse of IBD[7] and therefore should be employed with caution in patients with either ulcerative colitis or Crohn’s disease. A recent systematic review of the available medical literature concluded that the epidemiological evidence for a positive link between NSAID exposure and relapse of IBD is weak, while admitting that “some patients with IBD do relapse when given NSAIDs[8].

Given the inconsistency of the conflicting data concerning the relationship between NSAIDs and IBD, the possible effect of selective cyclooxygenase-2 inhibitors (COXIBs) in this respect remains even more controversial. In order to better understand the relationship between anti-inflammatory treatment and IBD it is necessary to consider the possible pathogenetic mechanisms involved in the adverse effects on the bowel by non-selective NSAIDs. Several mechanisms have been postulated, such as enhanced intestinal permeability[9], enterohepatic recirculation of NSAIDS and formation of drug enterocyte adducts , the latter phenomena having been observed in animal studies[9] but never demonstrated in humans.

The major mechanism involved, however, is thought to be the inhibition of colonic prostaglandin synthesis[10], in particular of the COX-2 isoform. In the inflamed colon COX-2 expression is upregulated in an effort to repair mucosal damage[11] and its inhibition may result in exacerbation of colonic injury and in impairment of the mucosal repair processes elicited by the COX-2 enzyme[12]. In this respect both NSAIDs and COX-2 inhibitors could hamper the progression of the inflammatory state toward healing. On the other hand, if COX-2 is important in the reparative mechanisms in IBD, then patients with quiescent disease should have a lower risk of flare-up when taking NSAIDs[13].

The studies on the effect of COX-2 inhibitors on animal models of colitis have yielded conflicting results[9,14] even taking in account the differences in experimental conditions, type and dosages of the employed compounds. The only available study on human colonic mucosa, carried out on colonic biopsies taken in IBD patients, found that a highly selective COX-2 inhibitor, L-745337 inhibits local release of PGE2 and PGI2 to the same extent as indomethacin, a nonselective NSAID[15], an effect which would likely promote aggravation of mucosal damage..

In a clinical setting a perspective, open-label study in IBD patients with associated arthropathy rofecoxib, administered at a dose of up to 25 mg daily for 20 d, failed to elicit any flare-up of the intestinal disease[16]. Similarly, a retrospective analysis of IBD patients treated with either celecoxib or rofecoxib for periods ranging from one week to 22 mo[17]. apparently confirmed the safety of COX-2 inhibitors in this respect. By contrast, a clinical exacerbation of the underlying IBD that subsided after the drug was discontinued, has been reported in 19% of patients taking rofecoxib[18]. In keeping with this finding a recent retrospective study in IBD patients taking either celecoxib or rofecoxib has found clinical relapse of the intestinal disease in 39% of cases, again with resolution of symptoms after COX-2 inhibitor withdrawal[19].

On the other hand, the first multicenter, random, double-blind, placebo-controlled study performed in USA ,taking into consideration of both clinical and endoscopic parameters, has shown that celecoxib 200 mg bid for 2 wk is as safe as placebo in patients with ulcerative colitis in remission[20].

Thus, as with nonselective NSAIDs, the available data remain conflicting and confusing. Summing up, on theoretical ground both NSAIDs and COX-2 inhibitors appear capable of triggering a flare-up of IBD by inhibiting the intestinal production of prostaglandins involved in the tissue reparative processes. In clinical practice, although clear-cut evidence is difficult to obtain due to the variable incidence of IBD reactivation and the paucity of prospective, controlled studies, both types of anti-inflammatory agents may precipitate recurrence of intestinal symptoms and therefore should be avoided, when possible, in patients with ulcerative colitis or Crohn’s disease.