Overall Study Question
To determine if eradication of Helicobacter pylori will reduce the risk of ulcers among patients positive for H pylori who have dyspepsia or a history of ulcer, and are about to start long-term non-steroidal anti-inflammatory drug (NSAID) treatment. All patients were NSAID naïve. The study sample included 100 arthritis patients, including 33 males and 67 females, with a mean age of 63 years. Sixteen patients had a history of ulcer, 60 patients had coexisting illness (cerebrovascular accident, ischemic heart disease, heart failure, chronic obstructive airway disease and chronic liver disease), and 17 used low dose ASA concurrently with NSAIDs.
The primary outcome variable was endoscopically proven gastric or duodenal ulcers. The secondary outcome was complicated ulcers – symptomatic and bleeding ulcers. The study intervention was triple therapy with oral omeprazole 20 mg, amoxicillin 1000 mg, and clarithromycin 500 mg each given twice daily for 1 week. The control group received oral omeprazole 20 mg, and placebo.
Are the Results of the Study Valid?
1. Was assignment of patients randomized?
2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?
Yes. One hundred of the 102 randomized patients (98%) were accounted for in the final analysis. The 2 patients not included in the final analysis withdrew from the placebo group prior to receiving any medication.
3. Were patients, their clinicians, and study personnel ‘blind’ to treatment?
Yes. The study investigators, study nurse, endoscopists, pathologists and patients were all blinded. Placebo medications were identical in appearance to study medications.
4. Were the groups similar at the start of the trial?
Yes. The two groups were similar for most of the reported baseline characteristics. The only potentially important difference was that the placebo group contained more smokers (10% vs. 18%).
5. Aside from the experimental intervention, were the groups treated equally?
Yes. Both groups were treated identically.
What were the Results?
1. How large was the treatment effect?
The 6-month probability of ulcers was 12.1% in the treatment group and 34.4% in the placebo group (log rank test, p=0.0085). The 6-month probability of complicated ulcers was 4.2% in the treatment group and 27.1% in the control group (log rank test, p=0.0026).
In the Cox’s proportional hazard model only, assignment to the placebo group was significantly associated with an increased risk of ulcers (hazard ratio 3.3 [95% CI 1.2-7.9]) and complicated ulcers (6.5 [95% CI 1.4-29.5]). Among patients with previous ulcer disease 1 of 7 in the treatment group and 5 of 9 in the placebo group had recurrent ulcers within 6 months (p=0.15). Among those who concurrently used low dose ASA, 1 of 9 in the treatment group and 2 of 8 in the placebo group developed ulcers within 6 months (p=0.58).
Finally, 43% of patients in the treatment group and 35% in the placebo group reported gastrointestinal upset during the first week of treatment (p=0.42).
2. How precise was the estimate of the treatment effect?
The 6-month probability of ulcers was 12.1% (95% CI 3.1-21.1) in the treatment group and 34.4% (95% CI 21.1-47.7) in the placebo group. Our calculation of the 95% confidence intervals for the ARR of 22.3% is 7%-38% (NNT 3-14).
Will the Results Help Me in Caring for My Patients?
1. Can the results be applied to my patient care?
Yes. This trial indicates that NSAID naive arthritis patients who are H pylori positive and have a history of dyspepsia or endoscopically proven peptic ulcer requiring long term regular NSAID treatment receive benefit from H pylori eradication begun concomitantly with NSAID treatment.
2. Were all clinically important outcomes considered?
The authors considered the probability of ulcers and complicated ulcers (bleeding or symptomatic) at 6 months. The Cox’s proportional hazard model was used to adjust for possible confounding covariates including age, coexisting illness, concurrent use of low dose aspirin, and previous ulcer disease. The authors could have considered ulcer disease beyond 6 months.
3. Are the likely treatment benefits worth the potential harms and costs?
A potential benefit of H pylori eradication is reduced dyspeptic symptoms. This small trial demonstrated a trend towards less dyspepsia in patients who received eradication therapy, as compared to those who received placebo. A potential harm of H pylori eradication is GI upset. In this study however, the relative incidences of GI upset during the first week of the study were similar between the eradication and placebo groups. The approximate drug acquisition cost for the H pylori eradication regimen used in this trial is ~$100-120 CAN. A decision analysis model has indicated screening and treatment of H pylori would be cost effective if it reduces ulcer risk by 50%.
There has been long standing controversy as to whether or not the presence of H pylori influences the risk of acquiring ulcers while on NSAID therapy. A recent meta-analysis has found that both H pylori and NSAID use are independent risk factors for developing peptic ulcer disease (PUD) and ulcer bleeding and that there is synergy in their ulcerogenic tendencies.
Logically, clinicians wonder if eradication of H pylori will decrease the incidence of PUD and bleeding in patients on NSAIDs. This study provides evidence that in older patients naïve to NSAIDs, eradiation of H pylori offers clinical benefit in reducing the incidence of ulcers and/or bleeding at 6 months. It also backs up the results of a previous study by these authors that addressed the same question in patients with no history of dyspepsia or ulcer. These results should be interpreted with caution however, because only 16% of patients had pre-existing peptic ulcer disease, an important risk factor for future ulcer or bleeding.
Whether this data can be extrapolated to other populations is unknown. It is apparent that the results cannot be extrapolated to those already on chronic NSAID treatment, as a previous study has shown that these patients do not have reduced ulcer risk from eradication of H pylori.
Another interesting question is whether COX-2 inhibitors are less likely to produce ulceration and bleeding in those infected with H pylori. The belief that COX-2 inhibitors cause less GI complications than traditional NSAIDs may tempt clinicians to advocate using a COX-2 inhibitor alone in H pylori infected patients, without first eradicating the H pylori. However, evidence suggests that COX-2 inhibitors also increase the ulcer risk of patients with H pylori infection. Therefore, a strategy of using COX-2 without eradicating H pylori can not be advocated at this time.
Finally, in patients with a history of peptic ulcer disease, or even dyspepsia (which is infrequently associated with ulcer disease), is it necessary to provide continual cytoprotection (e.g. misoprostol or a proton pump inhibitor) once H pylori has been eradicated? The ulcer rate at 6 months in the eradication group was lower, but was still 10%. This trial does not answer that question, although based on a review of other evidence, that strategy in patients with previous ulcer disease may be reasonable.
Chan FKL, To KF, Wu JCY, Yung MY, Leung WK, Kwok T, et al. Eradication of Helicobacter pylori and risk of peptic ulcers in patients starting long-term treatment with non-steroidal anti-inflammatory drugs: a randomised trial. Lancet 2002;359:9-13