Is test-and-eradicate or prompt endoscopy more effective for treatment of dyspepsia in Helicobacter pylori–positive patients?

Author and Disclosure Information

Lassen AT, Pedersen FM, Bytzer P, Schaffalitzky de Muckadell OB. Helicobacter pylori test-and-eradicate versus prompt endoscopy for management of dyspeptic patients: a randomized trial. Lancet 2000; 356:455-60.


BACKGROUND: A recent Cochrane review found that early endoscopy was not more effective than initial acid suppressive therapy, but the 3 available studies lacked statistical power to confirm a clear difference. No eligible trials of treatment guided by the results of H pylori testing versus treatment guided by initial endoscopy were found.1

POPULATION STUDIED: A total of 576 dyspeptic patients with a median age of approximately 45 years were recruited from Danish general practitioners. After exclusion for significant comorbidities or complicating factors, such as such as upper gastrointestinal bleeding, 500 patients were randomized, and 447 completed 1 year follow-up.

STUDY DESIGN AND VALIDITY: This was a randomized controlled trial of prompt endoscopy versus a test-and-eradicate strategy. Randomization using random number tables was accomplished using sealed numbered envelopes. Groups were similar at the beginning of the study, although patients in the prompt endoscopy group had a longer history of dyspepsia. Patients were interviewed at follow-up intervals of 1 month and 1 year and completed a diary for 1 week out of each month during the study year. Patients randomized to the test-and-eradicate strategy had C-urea breath testing for H pylori. Those with a positive test were treated with a 2-week course of lansoprazole 30 mg twice daily (a proton pump inhibitor [PPI]), metronidazole 500 mg 3 times daily, and amoxicillin 1000 mg twice daily. H pylori negative patients with primarily reflux symptoms were treated with a PPI for 1 month and continued on demand. Patients who did not improve or relapsed during follow-up were offered endoscopy. Patients randomized to prompt endoscopy were treated according to endoscopic findings. Those with duodenal ulcers received eradication therapy followed by 2 weeks of PPI. Patients with gastric ulcerations were treated according to H pylori status. Gastric ulcers were biopsied every 6 weeks until healed. Patients with reflux esophagitis were given PPI therapy, for 8 weeks of PPI. Those diagnosed with functional dyspepsia were reassured, given lifestyle modification advice, and prescribed PPIs if they had previously used them successfully. The strengths of this study include its randomized design, high follow-up rate, recruitment from primary care practices, and focus on patient-oriented outcomes. The authors do not mention if attempts were made to blind interviewers or assessors to patient group status. The results are likely applicable to primary care populations, but referral to a teaching hospital may have created expectation biases on the part of patients. Although analysis was not by intention to treat, the number of dropouts was the same between groups and was only 10% overall. The sample size was adequate to detect a difference of 10% in the number of days of symptoms.

OUTCOMES MEASURED: The main outcome was the percentage of days without dyspeptic symptoms. Secondary outcomes included the number of sick leave days, quality of life measured with a visual analog scale, and the gastrointestinal symptoms rating scale, visits to general practitioners, hospital admissions, satisfaction, use of endoscopy, and use of PPIs.

RESULTS: No differences were found in the proportion of days without dyspeptic symptoms, scores on quality of life measures, outpatient visits, hospital admissions, or sick leave days. Significantly more patients in the prompt endoscopy group (20% vs 28%) reported improved symptoms at 1 month, but there were no differences at 1 year. The patients assigned to the test-and-eradicate strategy were more likely to be dissatisfied at 1 year (12% vs 4%). As expected, the use of encoscopy (0.5 vs 1.25 per person, P <.001) was significantly lower in the test-and-eradicate group.


A test-and-eradicate strategy cured or identified 91% of all peptic ulcer patients and was as effective as immediate endoscopy at reducing the number of days of symptoms. In areas where access to endoscopy is institutionally limited, expensive, or there is a prolonged waiting time, a test-and-eradicate strategy is preferable. However, patient satisfaction monitoring should accompany adoption of this strategy. This is a rapidly evolving field, and several studies of endoscopy versus various treatments are due to be published soon.

Next Article: