Proton Pump Inhibitors to Prevent NSAID-Induced Ulcers and Dyspepsia

Are proton pump inhibitors (PPIs) effective for preventing dyspepsia and gastric or duodenal ulcers in people with long-term nonsteroidal anti-inflammatory drug (NSAID) consumption?

PPIs slightly reduce symptoms of dyspepsia and improve quality of life for people taking NSAIDs long-term compared with placebo. PPIs also decrease incident gastric and duodenal ulcers vs placebo. It is unclear if PPIs are effective for preventing gastric and duodenal ulcers compared with histamine H₂ blockers or misoprostol, or if there is any difference in adverse events compared with placebo or other drugs.¹ (Strength of Recommendation: B, inconsistent or limited-quality patient-oriented evidence.)

NSAIDs comprise 5% to 10% of all prescriptions and often are used long-term to treat inflammation or pain. Adverse effects of long-term NSAID use include gastrointestinal toxicity, ranging in severity from dyspepsia to gastric or duodenal ulcers, ulcer bleeding, and perforation. NSAIDs inhibit protective gastric prostaglandins and interfere with mucosal integrity.² PPIs provide gastroprotective effects by decreasing gastric acid secretion. The authors of this review aimed to assess the effects of PPIs on the prevention of dyspepsia and gastric or duodenal ulcers in people who used NSAIDs long-term (ie, at least 4 weeks).¹

This Cochrane review included 12 randomized controlled trials (N = 8,760) published between 1996 and 2014.¹ All studies were in the outpatient setting in Africa, Asia, Europe, North America, Central America, South America, and Australia. Studies included people taking various doses and kinds of NSAIDs, including cyclooxygenase-2 inhibitors and aspirin. Baseline ulcer risk was not reported. Participants were randomized to an oral PPI vs placebo (11 studies), an H₂ blocker (one study), or misoprostol (one study). The duration of follow-up was 12 months or less.

PPIs included esomeprazole, lansoprazole, omeprazole, and pantoprazole, given once daily (11 studies) or twice daily (one study), with four studies evaluating two different PPI daily dosages.¹ The main outcomes included global symptoms of dyspepsia (measured by a validated score or scale), incident ulcer (endoscopically confirmed gastric or duodenal ulcer), ulcer complications (bleeding or perforation), quality of life (measured by validated scales), and adverse events (any unexpected or harmful occurrence).

Compared with placebo, PPIs may slightly reduce global symptoms of dyspepsia on the Gastrointestinal Symptom Rating Scale (scale from 0–7; mean difference [MD] = −0.56; 95% CI, −0.74 to −0.38; two studies; n = 1,149; moderate-certainty evidence).¹ They may have little or no effect on symptoms measured as a dichotomous outcome (very low-certainty evidence). PPIs probably reduce incident ulcers compared with placebo (number needed to treat [NNT] to prevent one ulcer = 12 over 12–26 weeks; 11 studies; n = 7,022; moderate-certainty evidence). PPIs probably slightly increase quality of life (physical/social functioning, on a 7-point Likert scale) compared with placebo (MD = 0.39; 95% CI, 0.23–0.55; two studies; n = 1,149; moderate-certainty evidence). PPIs may have little to no effect on adverse events vs placebo (very low-certainty evidence).

One small study suggested PPIs may increase incident ulcers compared with H₂ blockers, but there were no other outcomes considered in the review (n = 26; low-certainty evidence).¹ Compared with misoprostol, PPIs may increase the risk for incident ulcer (number needed to harm to observe one additional ulcer = 9 over 12 weeks; one study; n = 402; very low-certainty evidence), but they were also much better tolerated, resulting in fewer adverse events compared with misoprostol (NNT to prevent one adverse event = 6; one study; n = 402; very low-certainty evidence).