Identifying and Treating Patients at Risk: Risk of NSAID-Related Gastrointestinal Injury

In: Main

29 Jun 2010

Gastrointestinal InjuryDuring the past two decades, numerous studies have quantified the relative risk for the development of gastrointestinal injury in patients receiving aspirin or nonselective COX-1/COX-2 NSAIDs.


Aspirin ingestion increases the relative risk of gastrointestinal injury by approximately threefold to fourfold. Virtually all healthy individuals exhibited gastric mucosal injury following aspirin ingestion. In a controlled study of 2,250 ulcer-free patients given aspirin 1 g/day for four years, the incidence of gastric and duodenal ulcers was nine to 11 times more prevalent among those receiving aspirin. In a meta-analysis of 21 primary and secondary myocardial and stroke-prevention trials, the Antiplatelet Trialists’ Collaboration found that the risk of significant gastrointestinal bleeding and hospitalization resulting from peptic ulceration was approximately twofold higher in patients treated with aspirin (in doses of 75-1,500 mg/day) compared with those receiving placebo. The odds ratios (ORs) for selected gastrointestinal symptoms and complications were as follows: all ulcers, 1.3; ulcers serious enough to lead to withdrawal, 3.2; all gastrointestinal symptoms, 1.7; and symptoms severe enough to cause a patient to withdraw, 1.5. canadian antibiotics

A meta-analysis of 24 randomized, controlled trials by Derry and Loke found that gastrointestinal hemorrhage occurred in 2.47% of individuals taking aspirin for one year compared with 1.42°% of patients taking placebo (OR = 1.68). At doses below 163 mg/day, gastrointestinal hemorrhage occurred in 2.3% of patients taking aspirin compared with 1.45% of those taking placebo (OR = 1.59). No relationship was observed between gastrointestinal hemorrhage and aspirin dose, although patients receiving modified-release aspirin formulations were at slightly lower risk (OR = 1.93).

These findings as well as those of others indicate that aspirin at any dose or formulation increases the risk of gastrointestinal injury and complications. In a study by Weil and colleagues, the odds ratios of upper gastrointestinal events by a daily dose of aspirin were as follows: 2.3 with a dose of 75 mg, 3.2 with a dose of 150 mg, and 3.9 with a dose of 300 mg. Kel-ley and associates found a similar risk (RR = 2.4-2.6) among those given plain aspirin, coated aspirin, or buffered aspirin. The relative risk of gastric bleeding was 2.6 with plain aspirin, 3.2 with coated aspirin, and 3.6 with buffered aspirin.

A recent article by Lai and colleagues found that initiation of antisecretory therapy (i.e., lansoprazole [TAP] 30 mg once daily) in combination with low-dose aspirin significantly reduced the risk of recurrent ulcer complications. The investigators enrolled 123 H. pylori-positive patients with ulcer complications after receiving low-dose aspirin for more than one month. Following ulcer healing and cure of the H. pylori infection, patients were randomly assigned to receive treatment with the proton pump inhibitor (PPI) or placebo in addition to aspirin 100 mg daily for 12 months. During the 12-month follow-up, 14.8% of aspirin/placebo-treated patients, in contrast to 1.6% of aspirin patients, experienced recurrence of ulcer complications (P = .008).

The aspirin/NSAIDs combination has a particularly deleterious effect on the gastric mucosa. A cohort study from Denmark found the incidence rate of gastrointestinal bleeding with low-dose aspirin to be 2.6 (95% confidence interval [CI], 2.2-2.9). Among patients who were also taking NSAIDs, the incidence rate rose to 5.6 (95% CI, 4.4-7.0). The increased gastrointestinal injury risk of aspirin plus NSAID therapy occurred whether or not patients were receiving COX-1/COX-2 or COX-2 selective inhibitors. Silverstein and colleagues found no significant differences in upper gastrointestinal ulcer complications, alone or combined with symptomatic ulcers, between patients receiving aspirin with a traditional COX-1/COX-2 NSAID and those receiving aspirin with celecoxib (Celebrex generic, Pharmacia).

COX-1 and COX-2 NSAIDs

Langman and colleagues found that the relative risk of bleeding peptic ulcers in patients receiving COX-1/COX-2 NSAIDs was 3.1, a risk similar to that observed in patients treated with aspirin (RR = 3.5). One study and a meta-analysis found that the overall estimated relative risk for hospitalization as a result of peptic ulceration was approximately 3 in patients receiving NSAIDs.

Ofman and colleagues assessed the risk of severe gastrointestinal complications (perforation, ulcers, and bleeding episodes [PUB]) in a meta-analysis of 13 NSAID-versus-placebo, randomized clinical trials; three previously unpublished Food and Drug Administration (FDA) placebo-controlled, randomized trials; nine cohort studies; and 23 case-control studies with data sufficiently clinically homogeneous to pool. The pooled odds ratio for PUB varied according to the study design and ranged from 2.7 to 5.4 from the nine cohort study of more than 750,000 person-years of exposure and the 16 NSAID-versus-placebo clinical trials, respectively.

COX-2 Selective Inhibitors

Clinical trial data suggest that the COX-2 selective NSAIDs, compared with nonselective COX-1/COX-2 NSAIDs, cause fewer serious gastrointestinal events; however, more recent case reports have cited serious gastrointestinal complications with the use of these agents. The relative safety of these agents, particularly in patients at high risk and in those requiring concurrent aspirin therapy, remains controversial.

Much of the debate regarding the safety of the selective COX-2 NSAIDs stems from the conflicting findings of two double-blind, randomized, prospective outcome studies, the Vioxx Gastrointestinal Outcomes Research (VIGOR) trial, which involved patients with rheumatoid arthritis, and the Long-Term Arthritis Safety Study (CLASS) trial, which involved patients with osteoarthritis (70%) and rheumatoid arthritis (30%). In the VIGOR trial, patients were monitored for a median of 9.0 months. Patients receiving rofecoxib 50 mg/day experienced significantly fewer upper gastrointestinal complications than those receiving naproxen 500 mg twice daily (cumulative incidence 2.1% vs. 4.5%, respectively; P = .005).

The differences observed between the COX-2 agent and may have been caused by several study-related factors. For example, the VIGOR trial included a low percentage (8%) of individuals considered at high risk (with a history of a clinical gastrointestinal event) for a gastrointestinal event, and patients taking concurrent aspirin were excluded. In contrast, the CLASS study by Silverstein and colleagues found no statistically significant difference in the annualized incidence rate of upper gastrointestinal complications between patients treated for six months with celecoxib 400 mg twice daily or conventional NSAIDS Qbuprofen 800 mg three times daily or twice daily) (0.76°% vs. 1.45°%, P = .09). A statistically significant difference was observed in the annualized incidence rates of upper gastrointestinal complications combined with symptomatic ulcers in patients treated with celecoxib versus conventional NSAIDs (2.08% vs. 3.54%, P = .02).

Unlike the patients in the VIGOR trial, patients taking aspirin therapy were not excluded from participating in the CLASS trial. In fact, when the percentage of patients with upper gastrointestinal ulcer complications, combined with symptomatic ulcers in patients using aspirin, was evaluated, no statistically significant differences were observed between patients receiving a traditional NSAID and patients receiving canadian celecoxib (2.12 vs. 2.01, P = .92, and 6.0 vs. 4.7, P = .49, respectively).

About this blog

Blog invites submissions of review articles, reports on clinical techniques, case reports, conference summaries, and articles of opinion pertinent to the control of pain and anxiety in dentistry.