Review: Misoprostol or COX-2–specific or selective NSAIDs reduce gastrointestinal complications and symptomatic ulcersPDF
ACP J Club. 2005 May-Jun;142:75. doi:10.7326/ACPJC-2005-142-3-075
Clinical Impact Ratings
Hooper L, Brown TJ, Elliott R, et al. The effectiveness of five strategies for the prevention of gastrointestinal toxicity induced by non-steroidal anti-inflammatory drugs: systematic review. BMJ. 2004;329:948. [PubMed ID: 15475342]
In patients taking nonselective nonsteroidal antiinflammatory drugs (NSAIDs), what are the effects of 5 gastroprotective strategies (NSAIDs plus H2-receptor antagonists, NSAIDs plus proton-pump inhibitors, NSAIDs plus misoprostol, cyclooxygenase-2 [COX-2]–selective NSAIDs alone, and COX-2–specific NSAIDs alone) for preventing gastrointestinal (GI) toxicity?
Data sources: 5 databases, bibliographies of relevant studies, and contacting authors.
Study selection and assessment: Randomized controlled trials (RCTs) ≥ 21 days in duration comparing 1 of the 5 gastroprotective strategies with nonselective NSAIDs alone on outcomes of GI toxicity. Quality assessment included randomization procedures, allocation concealment, blinding, and follow-up.
Outcomes: Primary outcomes included serious GI complications (e.g., hemorrhage, hemorrhagic erosions, recurrent upper GI bleeding, perforation, and pyloric obstruction) and symptomatic ulcers.
112 RCTs (n = 74 666) met the selection criteria. 15 RCTs (n = 2621) compared H2-receptor antagonists with placebo, 6 RCTs (n = 1358) compared proton-pump inhibitors with placebo, 23 RCTs (n = 16 945) compared misoprostol with placebo, 51 RCTs (n = 28 178) compared COX-2–selective drugs with nonselective NSAIDs, and 17 RCTs (n = 25 564) compared COX-2–specific drugs with nonselective NSAIDs. Effects of serious GI complications and symptomatic ulcers are in the Table.
In patients taking NSAIDs, misoprostol or COX-2–specific NSAIDs alone reduce serious gastrointestinal complications and symptomatic ulcers more than nonselective NSAIDs alone. Misoprostol, proton-pump inhibitors, and COX-2–selective and COX-2–specific NSAIDs reduce symptomatic ulcers more than NSAIDs alone.
Source of funding: NHS Executive.
For correspondence: Dr. L. Hooper, University of East Anglia, Norwich, England, UK. E-mail email@example.com.
Table. Effects of 5 gastroprotective strategies in patients taking nonsteroidal antiinflammatory drugs at ≥ 21 days*
|Outcomes||Number of trials (n)||Comparisons||Weighted event rates||RRR (95% CI)||NNT (CI)|
|Serious gastrointestinal complications||4 (894)||H2-receptor antagonist vs placebo||0% vs 0.25%||39% (−280 to 90)||Not significant|
|4 (1108)||Proton-pump inhibitor vs placebo||0.36% vs 0.46%||46% (−185 to 90)||Not significant|
|10 (11 507)||Misoprostol vs placebo||0.53% vs 0.86%||41% (7 to 62)||306 (161 to 3144)|
|19 (22 717)||COX-2–selective NSAIDs vs nonselective NSAIDs||0.18% vs 0.24%||34% (−15 to 63)||Not significant|
|11 (21 454)||COX-2–specific NSAIDs vs nonselective NSAIDs||0.45% vs 0.73%||45% (20 to 62)||354 (208 to 1200)|
|Symptomatic ulcers||2 (343)||H2-receptor antagonist vs placebo||0.48% vs 0%||25% (−89 to 1378)||Not significant|
|RRR (CI)||NNT (CI)|
|2 (343)||Proton-pump inhibitor vs placebo||1.1% vs 9.7%||91% (53 to 98)||12 (8 to 25)|
|2 (8913)||Misoprostol vs placebo||0.35% vs 0.89%||64% (33 to 80)||185 (119 to 426)|
|16 (21 371)||COX-2–selective NSAIDs vs nonselective NSAIDs||0.41% vs 0.57%||58% (35 to 73)||622 (266 to ∞ )|
|11 (21 722)||COX-2–specific NSAIDs vs nonselective NSAIDs||1.1% vs 1.8%||51% (38 to 62)||142 (91 to 316)|
*Abbreviations defined in Glossary; weighted event rates, RRR, RRI, NNT, NNH, and CI calculated from data in article using a random-effects model.
Several important conclusions can be drawn from this review by Hooper and colleagues. First, although the investigators found that COX-2–selective NSAIDs reduced the risk for symptomatic ulcers and gastrointestinal symptoms, no reductions were seen in endoscopic ulcers and ulcer complications. This finding raises concerns that COX-2–selective NSAIDs might actually do more harm than good by masking dyspeptic symptoms and should be avoided in high-risk patients. Second, although COX-2–specific NSAIDs, as a class, seem to reduce the risk for ulcer complications, only rofecoxib has been clearly shown to achieve this goal. Third, the strongest evidence for prevention of NSAID-induced ulcer complications was found with misoprostol, but its use is limited by other adverse effects. Although low-dose misoprostol is better tolerated, it fails to prevent ulcer complications (1). Fourth, no evidence exists from randomized, placebo-controlled trials that proton-pump inhibitors prevent ulcer complications as an adjunct to NSAID therapy.
This meta-analysis did not include direct comparisons of some of the treatment options. It is therefore unknown whether proton-pump inhibitors are more effective than double-dose H2-receptor antagonists or full-dose misoprostol (2). Although celecoxib was comparable to omeprazole and diclofenac for reducing the risk for recurrent ulcer bleeding (3), neither treatment could eliminate the hazard of ulcers in high-risk patients (4).
Vincent W.S. Wong, MBChB, MRCP
Francis K.L. Chan, MBChB, MD, FRCP, FACG
Chinese University of Hong Kong
Shatin, Hong Kong, China
1. Chan FK, Sung JJ, Ching JY, et al. Randomized trial of low-dose misoprostol and naproxen vs. nabumetone to prevent recurrent upper gastrointestinal haemorrhage in users of non-steroidal anti-inflammatory drugs. Aliment Pharmacol Ther. 2001;15:19-24. [PubMed ID: 11136274]
2. Graham DY. Critical effect of Helicobacter pylori infection on the effectiveness of omeprazole for prevention of gastric or duodenal ulcers among chronic NSAID users. Helicobacter. 2002;7:1-8. [PubMed ID: 11886469]
3. Chan FK, Hung LC, Suen BY, et al. Celecoxib versus diclofenac and omeprazole in reducing the risk of recurrent ulcer bleeding in patients with arthritis. N Engl J Med. 2002;347:2104-10. [PubMed ID: 12501222]
4. Chan FK, Hung LC, Suen BY, et al. Celecoxib versus diclofenac plus omeprazole in high-risk arthritis patients: results of a randomized double-blind trial. Gastroenterology. 2004;127:1038-43. [PubMed ID: 15480981]