Upper abdominal or epigastric pain, which is referred to as dyspepsia, may indicate the presence of peptic ulcer disease or gastric cancer. After diagnostic evaluation, most patients with dyspepsia have no anatomic or biochemical explanation for their pain. This condition is known as nonulcer dyspepsia and is usually treated with a histamine H₂-receptor blocker or a proton-pump inhibitor. The discovery of a causal link between Helicobacter pylori and peptic ulcer disease has led to frequent screening for this organism in patients with abdominal pain. Several studies have shown that about 30 percent of patients with nonulcer dyspepsia have H. pylori infection. Whether subsequent antibiotic treatment will benefit these patients is debatable. Published studies to date have shown conflicting results. Talley and colleagues conducted a randomized, placebo-controlled trial to determine the effect of eradicating H. pylori in patients with nonulcer dyspepsia.

Potentially eligible patients were 18 to 65 years of age and had at least a three-month history of dyspepsia. They also had moderate pain or discomfort in the upper abdomen for a minimum of three days during the week before randomization. Patients were excluded from the study if they had a history of gastroesophageal reflux, peptic ulcer disease or irritable bowel syndrome. Patients with dyspepsia were initially screened for H. pylori with a carbon-13 urea breath test. Patients who had positive test results were presumed to be infected with H. pylori. These patients underwent baseline endoscopy to exclude the possibility of the following: reflux esophagitis; Barrett's esophagus; esophageal, gastric or duodenal ulcerations; gastric cancer; or any other significant abnormalities.

Patients with normal endoscopy were randomly assigned to receive 20 mg of omeprazole, 1,000 mg of amoxicillin and 500 mg of clarithromycin twice daily for 14 days or twice-daily treatment with three identical placebo medications.

The patients had follow-up visits four to six weeks after treatment and again at three, six, nine and 12 months. Urea breath testing was done at the first and last visits. In addition, endoscopy with biopsy was repeated at the 12-month visit. If the breath test, the histopathology or both were positive, the patient was considered to be still infected with H. pylori.

All patients kept track of their dyspepsia symptoms with diary cards. At each follow-up visit, a scoring system called the Gastrointestinal Symptom Rating Scale was used to assess patient symptoms recorded during the preceding week. The scoring system is a validated 15-item tool that uses a graded symptom scale ranging from zero (no symptoms) to 6 (very severe symptoms). Treatment was considered to be successful if the patient reported no more than mild upper abdominal pain (a score of zero or 1) during the seven days before the 12-month visit. Treatment was considered to have failed if the patient had used medication for dyspepsia in the 30 days before the 12-month visit.

Initially, 640 patients were screened; eventually, 293 were randomized into the study. The mean age of all patients was 46 years, and the number of men and women was almost equal. At the first follow-up visit, 90 percent of patients in the treatment group were cured of H. pylori infection compared with 2 percent of patients in the placebo group. At 12 months, these numbers were 80 and 5 percent, respectively, based on breath testing and histopathology.

The treatment scores of the two groups were similar at all follow-up visits. Using an intention-to-treat analysis, 46 percent of patients in the treatment group had successful relief of their symptoms at the 12-month visit, compared with 50 percent of patients in the placebo group. When comparing the specific types of dyspepsia, which were described by patients as “ulcer-like,” “reflux-like” or “dysmotility-like,” the authors found little difference between the groups in the number of patients still having symptoms at 12 months. The mean weekly rate of antacid use was 5.7 tablets per week in the treatment group and 5.2 tablets in the placebo group. There was also no difference in the use of other gastrointestinal drugs, including H₂-receptor antagonists and proton-pump inhibitors, during the follow-up visits.

The authors conclude that treating H. pylori infection in patients with nonulcer dyspepsia does not produce significant improvement in symptoms.