Current issues of ACP Journal Club are published in Annals of Internal Medicine


Duodenal ulcer was healed solely by eradication of Helicobacter pylori

ACP J Club. 1994 July-Aug;121:1. doi:10.7326/ACPJC-1994-121-1-001

Source Citation

Hosking SW, Ling TK, Chung SC, et al. Duodenal ulcer healing by eradication of Helicobacter pylori without anti-acid treatment: randomised controlled trial. Lancet. 1994 Feb 26;343:508-10.



To determine if eradication of Helicobacter pylori alone, without acid suppression, is sufficient to heal duodenal ulcers.


5-week randomized, single-blind, controlled trial.


University hospital in Hong Kong.


160 patients (61% men) with dyspepsia and endoscopically proven duodenal ulcer who were between 16 and 75 years of age were eligible for inclusion. Exclusion criteria were gastrointestinal bleeding within the past 4 weeks, previous acid-reduction surgery, or pregnancy. 132 patients (83%) completed the study. 28 patients were excluded after entry because of absence of H. pylori (n = 7), poor compliance because of side effects (n = 10), drugs lost (n = 2), wrong diagnosis (n = 1), intercurrent illness (n = 1), and defaulted follow-up (n = 7).


Patients were assigned to receive a 1-week course of bismuth subcitrate, 120 mg; tetracycline, 500 mg; and metronidazole, 400 mg (BTM) 4 times daily (n = 80); or the same regimen plus omeprazole, 20 mg/d for 4 weeks (OBTM) (n = 80). Patients in the BTM group were given 60 antacid tablets to be taken as required. Patients recorded their symptoms in a diary and were examined after 1 week of treatment.

Main outcome measures

Rate of ulcer healing and eradication of H. pylori determined by endoscopy and biopsy, and relief of symptoms. Additionally, an intention-to-treat (ITT) analysis was done for all but those patients who had negative test results for H. pylori at the outset of the study.

Main results

The healing rate did not differ between patients receiving BTM and those receiving OBTM { P > 0.2}* (Table). Eradication of H. pylori also did not differ between groups { P = 0.16}* (Table). At the end of the first week of treatment, symptom relief was more common in patients receiving OBTM than in patients receiving BTM (2.9 vs 4.2 days of pain, P < 0.01). In an ITT analysis, no difference existed between treatment groups for rates of ulcer healing, eradication of H. pylori, or relief of symptoms.


Eradication of Helicobacter pylori allowed duodenal ulcer healing without additional acid suppression. The addition of omeprazole to the treatment regimen helped reduce ulcer pain more rapidly.

Source of funding: Not stated.

For article reprint: Dr. A.K.C. Li, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, Hong Kong. FAX 852-2603-6197.

*P values calculated from data in article.

Table. Bismuth subcitrate, tetracycline, and metronidazole (BTM) vs BTM plus omeprazole (OBTM) for duodenal ulcer†

Outcomes at 5 wk BTM OBTM RRI (95% CI) NNH
Healing rate 92.3% 94.0% 1.8% (-9 to 12) Not significant
Eradication of H. pylori 93.8% 98.5% 4.7% (-3 to 14) Not significant

†Abbreviations defined in Glossary; RRI, NNH, and CI calculated from data in article


No question now exists about the association of Helicobacter pylori with duodenal ulcers. The bacterium is present in > 90% of patients with duodenal ulcers, regardless of age. Further, it has been adequately shown that duodenal ulcers heal more rapidly and that patients seldom relapse after the organism is eradicated, leading to the occasional use of the word "cure" when discussing the disease (1). Although H. pylori may be necessary, it is not a sufficient single cause of duodenal ulcers (2). The one unfulfilled Koch postulate is the production of duodenal ulcers by new infection. Indirect evidence may be obtained by treating H. pylori infection only and observing the results. Hosking and colleagues have shown that eradication of this organism is associated with rapid healing in the absence of acid-suppressing drugs. They did not look at recurrence rates, and placebo controls were not used. Using a different protocol (and too few patients), Rauws and Tytgat (1) obtained similar results in 1990.

Because the weight of evidence that implicates H. pylori as a causal factor in duodenal ulcers has become quite convincing, the question now shifts to "who should be treated?" In view of the side effects and costs of the medication, this is not a simple question. The Consensus Development Conference of the National Institutes of Health in February 1994 recommended that all patients with duodenal ulcers, including those with first apparent occurrence (and all gastric ulcers) that are shown to be associated with H. pylori infection, be treated with antimicrobial agents (3). The results of the Hosking study support these recommendations.

James L. Achord, MD
University of Mississippi Medical CenterJackson, Mississippi, USA


1. Rauws EA, Tytgat GN. Cure of duodenal ulcer associated with eradication of Helicobacter pylori. Lancet. 1990;335:1233-5.

2. Feldman M.Helicobacter pylori and the etiology of duodenal ulcer: necessary but not sufficient. Am J Med. 1991;91:563-5.

3. NIH Consensus Conference.Helicobacter pylori in peptic ulcer disease. NIH Consensus Development Panel on Helicobacter pylori in Peptic Ulcer Disease. JAMA. 1994;272:65-9.