Stratifying the risk of NSAID-related upper gastrointestinal clinical events: results of a double-blind outcomes study in patients with rheumatoid arthritis

Gastrointestinal side-effects of traditional non-steroidal anti-inflammatory drugs and new formulations

Although adverse effects of non-steroidal anti-inflammatory drugs (NSAIDs) occur in only a small proportion of users, the widespread use of these drugs has resulted in a substantial overall number of affected persons who experience serious gastrointestinal complications. Dyspeptic symptoms are estimated to occur in 10-60% of NSAID users and lead to discontinuation of treatment in 5-15% of rheumatoid arthritis patients taking NSAIDs. It is now well established that the point prevalence of peptic ulcer disease in patients receiving conventional NSAID therapy ranges between 10 and 30%, representing a 10-30-fold increase over that found in the general population. One of 175 users of conventional NSAIDs in the USA will be hospitalized each year for NSAID-induced gastrointestinal damage. The mortality of hospitalized patients remains about 5-10%, with an expected annual death rate of 0.08%. The selective COX-II inhibitors (rofecoxib, celecoxib, parecoxib, etoricoxib, valdecoxib, lumiracoxib) show consistently comparable efficacy to that of conventional non-steroidal anti-inflammatory drugs (NSAIDs) in patients with rheumatoid arthritis and osteoarthritis, but have a significantly reduced propensity to cause gastrointestinal toxicity. In many cases, the gastric effects of therapeutically active doses of COX-II inhibitors are indistinguishable from placebo. The safety benefits of COX-2 inhibitors given alone appear similar to combined therapy with conventional NSAIDs and gastroprotective agents. These findings warrant the consideration of COX-II inhibitors as first-line therapy in patients requiring long-term pain control.

Challenges in managing NSAID-associated gastrointestinal tract injury

The use of non-selective non-steroidal anti-inflammatory drugs (NSAIDs) and/or aspirin is associated with upper gastrointestinal (UGI) tract injury, particularly in high-risk patients (e.g. those with prior upper GI ulcers and ulcer complications, individuals > or = 65 years, or those using high-dose or multiple non-selective NSAIDs). Important management issues regarding NSAID-associated upper gastrointestinal toxicity include: (1) the optimal therapeutic approach to accelerate the healing of gastroduodenal ulcers that may become clinically apparent while using NSAIDs and (2) approaches to prevent the development of gastroduodenal ulcers and ulcer complications associated with their continued use. Clinical trials have reproducibly demonstrated that the healing of NSAID-associated gastric and duodenal ulcers is accelerated with the use of acid suppressive agents (e.g. histamine-2-receptor antagonists, H(2)RAs, and proton pump inhibitors, PPIs), even with the continued use of the NSAID. The risk of developing gastroduodenal ulcers or ulcer complications with the continued and long-term use of NSAIDs is now well recognised as an important problem commonly encountered in daily clinical practice. Clinical trials have shown that co-prescription of misoprostol, high-dose H(2)RAs or PPIs can effectively prevent or reduce the rate of gastroduodenal mucosal damage associated with the use of non-selective NSAIDs. Approaching the problem in a different way, COX-2-selective inhibitors circumvent the problem; based on their mechanism of action, these agents are less ulcerogenic in UGI tract as compared with non-selective NSAIDs. In a recently reported trial conducted in high-risk patients (past history of a NSAID-associated UGI ulcer bleed), the use of a COX-2-selective inhibitor or the combination of non-selective NSAID with a PPI resulted in similar efficacy and both strategies reduced the risk of a recurrent UGI event as compared to historical controls. Aspirin is an independent risk factor for UGI tract injury, even at the low doses used for cardiovascular prophylaxis. While a tremendous amount of research supports the use of preventative therapies and interventions to reduce and /or avoid NSAID- or aspirin-associated ulcers and ulcer complications in the UGI tract, these strategies are often times under-utilized, sub- optimally dosed, and/or are associated with poor patient compliance. This reinforces the need for continued clinician and patient education to improve our outcomes of care.

Utilization of nonsteroidal anti-inflammatory drugs and antisecretory agents: a managed care claims analysis

PURPOSE

To describe patients initiating nonsteroidal anti-inflammatory drug (NSAID) therapy with regard to gastrointestinal and cardiac risks and patterns of antisecretory agent use, and to explore the relation between therapy type and subsequent outcomes.

METHODS

We studied patients aged 18 years or older who had continuous coverage from 1998 to 2001 and who had initiated treatment with cyclooxygenase-2 (COX-2) selective inhibitors or nonselective NSAIDs. Patients were categorized with respect to gastrointestinal and cardiac risk profiles. Proton pump inhibitor use within 15 days of initiating NSAID therapy was considered prophylactic. Logistic regression analysis was used to evaluate associations between treatment and hospitalization events, cardiac events, and health care costs.

RESULTS

We identified 106,564 eligible NSAID initiators: 65.2% used COX-2 inhibitors and 34.8% used traditional NSAIDs. Users of COX-2 inhibitors were more likely to be at higher risk of gastrointestinal bleeding and cardiac events than were NSAID users. Proton pump inhibitor prophylaxis was most common among users of COX-2 inhibitors, but was only 11% in patients at high risk of gastrointestinal bleeding. There were no differences among treatment groups in terms of gastrointestinal or cardiac events. Initiation of COX-2 inhibitor therapy was associated with greater total health care costs.

CONCLUSION

Although we found that COX-2 inhibitors were used more frequently than were traditional NSAIDs in certain groups of patients with varying cardiac or gastrointestinal risk, we did not find that their use resulted in reductions in clinical events, cotherapy with proton pump inhibitors, or costs, suggesting that a better understanding of the relation between NSAID treatment strategies and outcomes in patients with differing risk characteristics is needed.

Review article: prevention of non-steroidal anti-inflammatory drug gastrointestinal complications--review and recommendations based on risk assessment

The incidence of non-steroidal anti-inflammatory drug-related ulcer complications remains high despite the availability of potent anti-ulcer drugs and selective cyclo-oxygenase-2 inhibitors. Non-steroidal anti-inflammatory drug-related ulcer complications can be minimized by prospective assessment of patients' baseline risk, rational choice and use of non-steroidal anti-inflammatory drugs, and selective use of co-therapy strategies with gastroprotectives. Current recommendations regarding strategies using anti-ulcer drugs and cyclo-oxygenase-2 inhibitors for prevention of clinical non-steroidal anti-inflammatory drug upper gastrointestinal events are largely derived from studies using surrogates such as endoscopic ulcers, erosions, and symptoms in low- to average-risk patients. Conclusions based on surrogate and potentially manipulatable end-points are increasingly suspect with regard to applicability to clinical situations. This article reviews the risks associated with non-steroidal anti-inflammatory drugs including aspirin and includes the effect of the patients' baseline risk, and the confounding effects of Helicobacter pylori infection. In addition, uncertainties regarding the clinical efficacy of anti-ulcer drugs and cyclo-oxygenase-2 inhibitors against non-steroidal anti-inflammatory drug-related ulcer complications are put into perspective. We propose management strategies based on the risk category: low risk (absence of risk factors) (least ulcerogenic non-steroidal anti-inflammatory drug at lowest effective dose), moderate risk (one to two risk factors) (as above, plus an antisecretory agent or misoprostol or a cyclo-oxygenase-2 inhibitor), high risk (multiple risk factors or patients using concomitant low-dose aspirin, steroids, or anticoagulants) (cyclo-oxygenase-2 inhibitor alone with steroids, plus misoprostol with warfarin, or plus a proton pump inhibitors or misoprostol with aspirin), and very high risk (history of ulcer complications) (avoid all non-steroidal anti-inflammatory drugs, if possible or a cyclo-oxygenase-2 plus a proton pump inhibitors and/or misoprostol). The presence of H. pylori infection increases the risk of upper gastrointestinal complications in non-steroidal anti-inflammatory drug users by two- to fourfold suggesting that all patients requiring regular non-steroidal anti-inflammatory drug therapy be tested for H. pylori.

Épidémiologie des complications gastro-duodénales associées aux anti-inflammatoires non stéroïdiens

Nonsteroidal anti-inflammatory agents (NSAIDs) are among the most widely prescribed drugs worldwide. In France, about 25% of individuals aged 40 years or older are treated with NSAIDs at least one week in the year. Although the therapeutic benefits of these drugs are substantial, their use is limited by their gastroduodenal toxicity. Dyspepsia occurs in about 30% of patients receiving NSAIDs, an approximately two-fold enhancement of risk compared with control subjects. Asymptomatic endoscopic lesions are observed in 20 to 80% of patients, depending on population characteristics, individual NSAIDs and definitions of endoscopic lesions. The risk of symptomatic ulcer, complicated ulcer (haemorrhage, perforation, stenosis) and death related to ulcer complication is multiplied by 4 in patients treated with NSAIDs. Established risk factors for NSAID-induced gastroduodenal complications are age, ulcer history, heavy alcohol consumption, individual NSAIDs, dose, association with corticoid or aspirin or anticoagulants (ulcer haemorrhage) while the role of treatment duration and Helicobacter pylori infection are controversial.

COX-2-selective inhibitors and the risk of upper gastrointestinal bleeding in high-risk patients with previous gastrointestinal diseases: a population-based case-control study

BACKGROUND

Clinical trials have suggested that cyclo-oxygenase-2-selective inhibitors are associated with a lower risk of upper gastrointestinal bleeding than are non-selective, non-aspirin, non-steroidal anti-inflammatory drugs. This has not yet been confirmed in studies of patients with an increased susceptibility to upper gastrointestinal bleeding.

AIM

To examine the risk of upper gastrointestinal bleeding in high-risk patients who filled prescriptions for cyclo-oxygenase-2 inhibitors or other non-steroidal anti-inflammatory drugs.

METHODS

A population-based case-control study was performed in the Danish county of North Jutland from 1 January 2000 to 31 December 2002. From the County Hospital Discharge Registry and the Civil Registration System, we identified incident cases with upper gastrointestinal bleeding (n = 780) and randomly selected controls (n = 2906), respectively. All cases and controls had previous gastrointestinal diseases. Data on drug exposure were obtained from the countywide Prescription Database.

RESULTS

Thirty-five cases (4.5%) filled prescriptions for cyclo-oxygenase-2 inhibitors within 30 days of the date of upper gastrointestinal bleeding, compared with 79 controls (2.7%). Adjusted odds ratios for upper gastrointestinal bleeding according to prescription for celecoxib, rofecoxib and non-steroidal anti-inflammatory drugs were 1.3 [95% confidence interval (CI), 0.7-2.8], 2.1 (95% CI, 1.2-3.5) and 3.3 (95% CI, 2.4-4.4), respectively.

CONCLUSIONS

In patients with increased susceptibility to gastrointestinal adverse events, a lower risk of upper gastrointestinal bleeding was observed in users of cyclo-oxygenase-2 inhibitors compared with users of other non-aspirin, non-steroidal anti-inflammatory drugs.

Inhibiteurs sélectifs de COX-2 : quelle toxicité gastro-duodénale ?

Endoscopic studies have shown that cyclo-oxygenase-1 sparing non steroidal anti-inflammatory drugs (NSAIDs) of the coxib family did not produce more gastroduodenal ulcers than placebo in patients who were free of ulcer at baseline. However, the clinical relevance of these data is disputable. Only celecoxib and rofecoxib have been subject to large scale gastrointestinal outcome studies. Both have been shown to reduce the risk of symptomatic ulcers and ulcer complications (perforation, gastric outlet obstruction, bleeding) by about 50% compared to classical NSAIDs used at their maximum therapeutic doses. However, coxibs appear to retain some residual risk, especially in patients at high risk of developing serious gastrointestinal adverse events. This is the case in patients requiring low dose aspirin for cardiovascular purposes. Whether coxibs are still less toxic to the gastroduodenal tract that non-selective NSAIDs in these patients is controversial. Although coxibs have a superior upper gastrointestinal tolerability relative to traditional NSAIDs, dyspepsia, abdominal pain and nausea remain the major factor limiting their use.

Impact de l’infection à Helicobacter pylori sur le risque de complications gastro-duodénales des traitements anti-inflammatoires non stéroïdiens

The interaction of Helicobacter pylori (H. pylori) and non steroidal anti-inflammatory drugs (NSAIDs) on the development of gastro-duodenal ulcers and their complications is complex and controversial. From a clinical point of view, the question is whether or not H. pylori infection should be tested and eradicated in patients treated or about to be treated by NSAIDs or low-dose aspirin. Contradictory results have been reported in epidemiological studies. Recent data suggest that H. pylori-NSAID interaction may be different depending on the type of treatment, non aspirin NSAIDs or low-dose aspirin, the gastric or duodenal localization of ulcer and the strains of H. pylori. Controlled randomized studies suggest that eradication of H. pylori may be beneficial in NSAID-naïve patients but not in those already on long term NSAID therapy. Recommendations are proposed for different subgroups of patients. In NSAID users presenting with gastro-duodenal ulcer or complications, H. pylori screening and eradication are indicated. In patients treated or about to be treated by NSAIDs, the "test and treat" H. pylori strategy is recommended if there is a history of gastroduodenal ulcer or complications. Whether this strategy should be generalized preventively in patients without ulcer history is still controversial and deserves further studies.

Randomized, Double-Blind, Placebo-Controlled Phase IIB Trial of the Cyclooxygenase Inhibitor Ketorolac as an Oral Rinse in Oropharyngeal Leukoplakia

PURPOSE

Nonselective cyclooxygenase (COX) inhibitors have been reported to decrease the frequency of upper aerodigestive cancers. Ketorolac tromethamine oral rinse has been shown to resolve another COX-dependent process, periodontal disease, without incurring gastrointestinal side effects. This trial evaluated if a topically delivered oral rinse containing ketorolac was as safe as and more effective than oral rinse alone in reducing the area of oral leukoplakia.

EXPERIMENTAL DESIGN

57 patients were randomized (2:1 ratio) in a double-blind, placebo-controlled study of ketorolac (10 ml of a 0.1% ketorolac rinse solution; n = 38) or placebo (10 ml of rinse solution; n = 19) given twice daily for 30 s over 90 days. Primary end point was evaluated visually obtaining bidimensional measurement of the size of leukoplakia lesion(s) at entry and at 90 days. Secondary end point was histological assessment of the leukoplakia as sampled by serial punch biopsy and independently reviewed by three pathologists.

RESULTS

The patients included 67% males, 11% non-Caucasian, and 86% used tobacco with no significant differences between the two arms. Both rinses were well tolerated with good compliance, and there was no significant difference in adverse events (P = 0.27). Major response rate (complete response and partial response) was 30% for ketorolac and 32% for the placebo arm. There was no significant difference in change in histology between the two arms.

CONCLUSION

Local delivery of a COX-containing oral rinse was well tolerated but produced no significant reduction in the extent of leukoplakia compared with the placebo. However, the favorable response rate to placebo arm remains unexplained and additional investigation of the tissue penetration with ketorolac is warranted.

Review article: appropriate use of proton pump inhibitors with traditional nonsteroidal anti-inflammatory drugs and COX-2 selective inhibitors

Nonsteroidal anti-inflammatory drugs (NSAIDs) are one of the most commonly used classes of medications in the USA, annually accounting for over 100 million prescriptions. Gastrointestinal complications associated with NSAIDs are common, and result in a substantial amount of morbidity and mortality, despite the advent of the cyclooxygenase-2 selective inhibitors or 'coxibs'. Emerging clinical and economic data suggest that, depending on the baseline risk to patients, the use of a traditional NSAID alone or in combination with a proton pump inhibitor are effective and well tolerated alternatives to coxibs. The optimal therapeutic strategy for NSAID selection and use of co-therapy should be guided by a consideration of each patient's risk of having an adverse event arising from the NSAID. Patients at the highest risk for gastrointestinal complications with traditional NSAIDs are those with a history of an ulcer or ulcer complication, those of advanced age (greater than 65 years), and those receiving concurrent aspirin, anticoagulants or corticosteroid therapy. Proton pump inhibitor co-therapy is highly effective in reducing NSAID-related dyspeptic symptoms, healing the injured mucosa even in those who continue to ingest NSAIDs, and preventing gastrointestinal complications. In addition to their selective use in patients who experience NSAID-related dyspepsia and other symptoms, proton pump inhibitor co-therapy should be considered in those at high risk (with coxib or traditional NSAID therapy) and is necessary in high-risk patients receiving aspirin, with or without NSAID therapy.

Single dose oral diclofenac for postoperative pain

BACKGROUND

Diclofenac is a benzene-acetic acid derivative that acts, like other NSAIDs, by inhibiting cyclo-oxygenase isoforms that mediate the body's production of the prostaglandins implicated in pain and inflammation. Diclofenac is widely available as a sodium or potassium salt. Diclofenac potassium tablets are known as 'immediate-release' diclofenac as absorption takes place in the gastrointestinal tract whereas 'delayed-release' (enteric-coated) diclofenac tablets resist dissolution until reaching the duodenum. An existing review showed that diclofenac was an effective treatment for acute postoperative pain but did not address the distinction between potassium and sodium salts due to lack of data. The aim of this update is to gather and add appropriate information published subsequently and, data permitting, examine any potential differences between the two different diclofenac formulations.

OBJECTIVES

To assess single dose oral diclofenac for the treatment of acute postoperative pain and determine whether there are differences between the different formulations.

SEARCH STRATEGY

We searched the Cochrane Library (Issue 2, 2003), MEDLINE (1966 to May 1996), EMBASE (1980 to 1996), Biological Abstracts (1985 to 2003), the Oxford Pain Relief Database (1950 to 1994), PubMed (1996 to 2003) and reference lists of articles.

SELECTION CRITERIA

Randomised, double-blind, placebo-controlled clinical trials of single dose, oral diclofenac sodium or diclofenac potassium for acute postoperative pain in adults.

DATA COLLECTION AND ANALYSIS

Two reviewers independently assessed trials for inclusion in the review, quality and extracted data. The area under the pain relief versus time curve was used to derive the proportion of patients prescribed diclofenac or placebo with at least 50% pain relief over four to six hours using validated equations. The number needed to treat (NNT) was calculated. Information on adverse effects was also collected.

MAIN RESULTS

One additional trial was included and added to the six trials included in the original review. All seven trials provided data for quantitative analysis: 581 patients were treated with diclofenac and 364 were treated with placebo. The NNT for at least 50% relief over four to six hours with diclofenac 25 mg, 50 mg and 100 mg compared with placebo was 2.8 (95% CI 2.1 to 4.3), 2.3 (2.0 to 2.7) and 1.9 (1.6 to 2.2) respectively. Though higher doses produced lower (better) NNTs, statistical significance was not achieved. There was no significant difference between diclofenac 50 mg and placebo in the proportion of patients experiencing dizziness, headache, nausea or vomiting. The weighted median duration of analgesia was 2 hours for placebo, 6.7 hours for diclofenac 50 mg and 7.2 hours for diclofenac 100 mg. Sensitivity analyses for drug formulation, pain model, trial size and quality did not reveal any statistically significant differences.

REVIEWERS' CONCLUSIONS

Oral diclofenac is an effective single-dose treatment for moderate to severe postoperative pain. There was no significant difference between diclofenac and placebo in the incidence of adverse effects, or between diclofenac sodium and potassium, different pain models, smaller and larger trials and trials of higher and lower quality.

Adherence to proton pump inhibitors or H2-receptor antagonists during the use of non-steroidal anti-inflammatory drugs

BACKGROUND

The efficacy of proton pump inhibitors (PPIs) or histamine-2 receptor antagonists (H2RAs) prescribed as prophylaxis for NSAID-related upper gastrointestinal (UGI) toxicity is dependent upon patient adherence.

AIM

To describe patient adherence to prophylactically prescribed PPIs and H2RAs in the clinical setting.

METHODS

We conducted a retrospective observational cohort study using the Integrated Primary Care Information Project database. The study population consisted of incident non-specific NSAID users prescribed a PPI or H2RA specifically as prophylaxis for NSAID-related UGI toxicity. Patients were classified as non-adherent if < 75% of days of NSAID use were covered by one of these agents, and as continuing users after discontinuation of NSAID use if they had a renewed prescription for these agents after their last NSAID prescription.

RESULTS

The study cohort comprised 784 patients: 374 with H2RAs, 405 with PPIs, and 5 with both PPI and H2RA. Eighty-five percent of H2RA users and 7% of PPI users were prescribed these drugs at doses below the minimum recommended/effective dose for NSAID-associated gastroduodenal ulcer prophylaxis. Thirty-seven percent of patients were non-adherent. The lowest rate of non-adherence was associated with the first NSAID prescription (9%), increasing to 61% for patients with >/= 3 prescriptions. In a cohort of subjects who stopped their NSAID and were followed for up to 2 years (n = 711), there was significant persistent use of acid suppressive agents; 40% of patients had at least one additional prescription for the acid suppressive agent after stopping NSAIDs, and> 30% received enough drug to cover a period longer than 2 months after stopping their NSAID.

CONCLUSIONS

The pattern of PPI and H2RA prescriptions, when prescribed as prophylactic strategy, does not correspond with the pattern of NSAID use. Physicians should consider the medical impact of non-adherence with dual therapies and the impact of prolonged use of GPAs on treatment cost.

Assessment of the safety of selective cyclo-oxygenase-2 inhibitors: where are we in 2003?

Non-steroidal anti-inflammatory drugs (NSAIDs) are among the most widely used drugs worldwide despite their well-documented adverse gastrointestinal (GI) effects. The risk of developing a severe GI event varies from patient to patient and NSAID to NSAID. Selective cyclo-oxygenase-2 inhibitors (coxibs) have been designed to have similar efficacy but less GI toxicity than traditional NSAIDs, and have been shown to have an improved GI tolerability and less adverse events across a range of different GI safety assessments. In clinical trials, particularly VIGOR and CLASS, rofecoxib and celecoxib, respectively, significantly reduce the risk of ulcers and ulcer complications than nonselective NSAID comparators with ulcer rates comparable to placebo. The real benefit of a coxib comes from the sparing of the thromboxane and hence preservation of normal platelet function. Thus, there is less risk of bleeding with selective inhibition of COX-2, which is the most common and serious complication of non-selective NSAIDs. Moreover, bleeding can occur anywhere in the GI tract. Although some concern has been raised about the cardiovascular safety of coxibs, when used in recommended doses, there is no convincing evidence that patients treated with a coxib have an increased risk of cardiovascular thrombotic events. Different approaches have been advocated to minimize NSAID-related GI toxicity. Choice of less harmful NSAIDs such as coxib has been one of the strategies promoted in guidelines. The introduction of coxibs with a higher benefit-risk ratio has dramatically changed the therapeutic scenario for anti-inflammatory treatment in the clinical practice.

Gastroprotection by coxibs: what do the Celecoxib Long-Term Arthritis Safety Study and the Vioxx Gastrointestinal Outcomes Research Trial tell us?

Current evidence suggests that PPIs might be effective in maintaining patients in remission during continued NSAID use and that the combination of omeprazole plus diclofenac is as effective as treatment with celecoxib in preventing recurrent bleeding. Larger outcome studies comparing the combination of a PPI with other nonselective NSAIDs and a selective COX-2 inhibitor (and the combination of a selective COX-2 inhibitor with a PPI or misoprostol) are required to determine whether or not any regimen will further decrease or eliminate the risk of ulcer complications in high-risk individuals.

The management of peptic ulcer disease

PURPOSE OF REVIEW

The period under review has seen little evolution in our understanding of the empiric management of dyspepsia. The role of Helicobacter pylori in this setting remains controversial, and a policy of risk stratification with the prudent use of test and treat and symptomatic therapy, with endoscopy for nonresponsive cases, seems to have some support from the literature in this period.

RECENT FINDINGS

The management of nonsteroidal antiinflammatory drug-associated and aspirin-associated complications has received a lot of attention in the period under review. The COX-2 selective agents have maintained their reputation as safer (but not "safe") options, although some of the original work with one of these agents has been rigorously interrogated and found wanting. Studies in the review period have focused our attention on the less than satisfactory protection of proton pump inhibitor cotherapy, the site-specific nature of ulcer recurrences (which may have therapeutic implications), lower gastroenterology complications associated with NSAID use, and the beneficial effect of proton pump inhibitor cotherapy for patients receiving low-dose aspirin. One should also expect a lot more information in the future with regard to the use of the nitric oxide donating class of nonsteroidal antiinflammatory drugs and aspirin.

SUMMARY

Findings are presented that suggest that the H.pylori stool antigen test is not as reliable as the urea breath test, while the most promising "new therapy" for H. pylori is not new, but rather an amalgam of some older drugs combined in a new "quadruple" therapy strategy, which shows some promise.

A comparison of the upper gastrointestinal mucosal effects of valdecoxib, naproxen and placebo in healthy elderly subjects

BACKGROUND

In long-term outcomes studies, cyclooxygenase COX-2 specific inhibitors spare COX-1 at supratherapeutic doses and therefore demonstrate improved gastrointestinal safety over nonspecific nonsteroidal anti-inflammatory drugs (NSAIDs). However, in clinical practice, anti-inflammatory drugs are often used for short-term treatment of pain.

AIM

To compare the short-term upper gastrointestinal mucosal effects of naproxen with the new COX-2 specific inhibitor, valdecoxib, or placebo, in elderly subjects.

METHODS

In this multicentre, double-blind, randomized, study, elderly subjects (65-76 years old), with a normal baseline esophagogastroduodenoscopy (EGD), received oral valdecoxib (a supratherapeutic 40 mg b.d. dosage, n = 62), naproxen (500 mg b.d., n = 62), or placebo (n = 62) for 6.5 days. Upper gastrointestinal mucosal injury was evaluated post-treatment by EGD (day 7).

RESULTS

Subjects receiving naproxen (11/60, 18%) had significantly more gastroduodenal ulcers post-treatment than those receiving placebo (2/61, 3%; P < 0.01) or valdecoxib (0/60, 0%; P < 0.001). A similar significant finding was observed for gastric ulcer rates. All treatments had similar adverse event rates and clinical laboratory findings.

CONCLUSIONS

Valdecoxib, even at supratherapeutic doses, was associated with an ulcer rate significantly lower than naproxen but similar to placebo in healthy elderly subjects, despite the short duration of therapy (6.5 days). Naproxen and valdecoxib were as well tolerated as placebo.

Effect of cyclooxygenase-2 inhibition on human Helicobacter pylori gastritis: mechanisms underlying gastrointestinal safety and implications for cancer chemoprevention

Cyclooxygenase (COX)-2 expression and prostaglandin production is increased by Helicobacter pylori infection. Non-selective COX inhibitors reduce prostaglandins and mucosal proliferation in infected mucosa and may reduce gastric cancer risk, but ulceration precludes their use. COX-2 inhibitors cause fewer ulcers and may be chemopreventive. Physiological studies of COX-2 inhibitors in humans with H. pylori infection have not been performed.

AIM

To study the impact of COX-2 specific inhibition on gastric prostaglandin levels, H. pylori gastritis and proliferation.

METHODS

Twenty infected (eight males, 12 females; age 38 +/- 1.8) and six uninfected (four males, two females; age 36 +/- 3.5) healthy volunteers received rofecoxib 25 mg daily for 14 days. Endoscopic biopsies were evaluated for prostaglandin E2 (PGE2) content, gastritis and proliferation.

RESULTS

Before drug therapy, compared to uninfected, H. pylori-infected subjects had significantly higher: (a) gastric mucosal PGE2 (pg/mg tissue) in the gastric body and antrum, (b) H. pylori score in body and antrum and (c) mid-gland proliferation index in antrum and body. The COX-2 inhibitor did not significantly affect PGE2 levels, gastritis scores or proliferation indices in the body or antrum in the H. pylori-positive or -negative subjects.

CONCLUSION

The predominant source of increased gastric PGE2 in H. pylori infection appears to be COX-1-derived. In non-ulcerated H. pylori gastritis, COX-2 inhibition does not affect cellular proliferation. Rofecoxib's lack of effect on gastric prostaglandin levels and proliferation in H. pylori-infected mucosa may explain the absence of an increased ulcer risk among COX-2 inhibitor users with H. pylori infection. The lack of significant effect on intermediate biomarkers raises uncertainty regarding the potential of specific COX-2 inhibitors for chemoprevention of gastric cancer.

Review article: NSAIDs, gastroprotection and cyclo-oxygenase-II-selective inhibitors

In patients at high risk of NSAID-associated serious upper gastrointestinal complications, gastroprotection with misoprostol or a proton pump inhibitor should be considered. Only misoprostol, 800 micro g/day, has been shown to reduce serious upper gastrointestinal complications in a large clinical outcome trial. The benefit of Helicobacter pylori eradication in reducing NSAID-associated gastrointestinal toxicity is controversial, and routine testing for and eradication of H. pylori in NSAID users are not currently advised. The gastrointestinal safety of rofecoxib and celecoxib has been assessed in large clinical outcome trials which, on first analysis, show benefits over non-selective NSAIDs in the incidence of serious upper gastrointestinal complications. However, longer term gastrointestinal data from the celecoxib study (CLASS) and cardiovascular adverse event data from the rofecoxib study (VIGOR) have questioned the risk-benefit profile of these new drugs and, until they are better understood, it seems sensible not to use them routinely in large numbers of individuals. The gastrointestinal safety of meloxicam and etodolac has not been adequately assessed in such trials. Therefore, evidence for their use instead of non-selective NSAIDs, or instead of celecoxib or rofecoxib, is not robust.