Elsevier

The Lancet

Volume 359, Issue 9300, 5 January 2002, Pages 9-13
The Lancet

Articles
Eradication of Helicobacter pylori and risk of peptic ulcers in patients starting long-term treatment with non-steroidal anti-inflammatory drugs: a randomised trial

https://doi.org/10.1016/S0140-6736(02)07272-0Get rights and content

Summary

Background

Whether Helicobacter pylori increases the risk of ulcers in patients taking non-steroidal anti-inflammatory drugs (NSAIDs) is controversial. We hypothesised that eradication of H pylori infection would reduce the risk of ulcers for patients starting long-term NSAID treatment.

Methods

Patients were enrolled if they were NSAID naïve, had a positive urea breath test, had dyspepsia or an ulcer history, and required long-term NSAID treatment. They were randomly assigned omeprazole triple therapy (eradication group) or omeprazole with placebo antibiotics (placebo group) for 1 week. All patients were given diclofenac slow release 100 mg daily for 6 months from randomisation. Endoscopy was done at 6 months or if severe dyspepsia or gastrointestinal bleeding occurred. The primary endpoint was the probability of ulcers within 6 months. Analyses were by intention to treat.

Findings

Of 210 arthritis patients screened, 128 (61%) were positive for H pylori. 102 patients were enrolled, and 100 were included in the intention-to-treat analysis. H pylori was eradicated in 90% of the eradication group and 6% of the placebo group. Five of 51 eradication-group patients and 15 of 49 placebo-group patients had ulcers. The 6-month probability of ulcers was 12·1% (95% CI 3·1–21·1) in the eradication group and 34·4% (21·1–47·7) in the placebo group (p=0·0085). The corresponding 6-month probabilities of complicated ulcers were 4·2% (1·3–9·7) and 27·1% (14·7–39·5; p=0·0026).

Interpretation

Screening and treatment for H pylori infection significantly reduces the risk of ulcers for patients starting long-term NSAID treatment.

Introduction

Peptic ulcers associated with use of non-steroidal antiinflammatory drugs (NSAIDs) are a serious public-health problem. Current strategies to prevent ulcer disease in NSAID users include concomitant therapy with anti-ulcer drugs or the use of NSAIDs selective for cyclo-oxygenase 2.1, 2, 3 Such approaches impose a heavy burden on health-care resources.

One other strategy to prevent NSAID-associated ulcer disease is to eliminate reversible risk factors, one of which might be Helicobacter pylori.4, 5 This bacterium is widespread; for example, it is present in 50% of the US non-white population older than 60 years.6 Studies to assess whether eradication of H pylori reduces the risk of ulcers in NSAID users have had conflicting results. Our earlier study showed that for patients who had no previous exposure to NSAIDs, eradication of H pylori with bismuth triple therapy before NSAID treatment reduced the risk of ulcers by 73% in 2 months.7 By contrast, for patients who are already receiving long-term NSAID treatment, cure of H pylori infection alone does not decrease gastroduodenal injury8 or adequately prevent recurrent ulcer bleeding.9 These divergent findings suggest that patients who have no previous exposure to NSAIDs differ from those who are already receiving these drugs. Epidemiological studies show that the risk of ulcers is two to four times higher among patients who have recently started NSAID treatment than among long-term users.10, 11, 12, 13 From a clinical point of view, the important question is whether patients without previous exposure to NSAIDs, who are commonly seen by primary-care physicians, are at increased risk of developing ulcer disease associated with NSAIDs if they are infected with H pylori.

The benefits of H pylori eradication for patients who are about to start long-term NSAID treatment remain uncertain, however, because our previous study had various limitations: follow-up of only 2 months; inclusion of only low-risk patients with no dyspepsia or ulcers; use of bismuth (which might have cytoprotective effects); exclusion of patients with active ulcers by screening endoscopy; and use of symptomless ulcers as the endpoint.7 In this study, we tested the hypothesis that among patients positive for H pylori who have dyspepsia or a history of ulcer and who are about to start long-term NSAID treatment, eradication of H pylori infection will reduce the risk of ulcers. We replaced bismuth by a proton-pump inhibitor in the eradication regimen and endoscopy by urea breath test, lengthened the observation period to 6 months, and assessed the frequency of both complicated and symptomless ulcers.

Section snippets

Patients

The protocol of this 6-month, double-blind, randomised, placebo-controlled trial was approved by the Ethics Committee of the Medical Faculty at the Chinese University of Hong Kong. The investigators screened arthritis patients who attended the Family Clinic or General Outpatient Clinic with informed written consent. Patients were enrolled if they required long-term regular NSAID treatment; had a positive urea breath test for H pylori; and had moderate dyspepsia or a history of endoscopically

Results

We screened 210 potentially eligible patients between March, 1999, and May, 2000. 128 (61%) patients were positive for H pylori infection, and 102 of these agreed to take part in the study (figure 1). Two patients assigned to the placebo group, who withdrew consent after randomisation and did not take any study medication, were excluded from the analysis. 100 patients (51 in the eradication group, 49 in the placebo group) were included in the intention-to-treat analysis. The two treatment

Discussion

This study shows that, among patients positive for H pylori who have dyspepsia or a history of ulcer and start long-term NSAID treatment, eradication of H pylori reduces the risk of endoscopic and complicated ulcers.

Epidemiological studies have consistently shown that the risk of ulcer complications is substantially increased during the first few months of NSAID treatment.10, 11, 12, 13 We postulate that this excess risk occurs in a subgroup of patients positive for H pylori who are prone to

References (26)

  • RI Russell

    Helicobacter pylori and non-steroidal anti-inflammatory drugs: ulcers and bleeding ulcers

    Ital J Gastroenterol Hepatol

    (1999)
  • DM McCarthy

    Helicobacter pylori and non-steroidal antiinflammatory drugs: does infection affect the outcome of NSAID therapy?

    Yale J Biol Med

    (1998)
  • JE Everhart et al.

    Seroprevalence and ethnic differences in Helicobacter pylori infection among adults in the Unites States

    J Infect Dis

    (2000)
  • Cited by (0)

    View full text