PREVENTION OF DUODENAL-ULCER RECURRENCE WITH PENICILLIN - A DOUBLE-BLIND, PLACEBO-CONTROLLED TRIAL

Eradication of Helicobacter pylori is associated with a reduced recurr ence of duodenal ulcer (DU). The relationship between H. pylori and DU has been interpreted as causal, but the evidence has been criticized for methodologic reasons. To ascertain whether an antibiotic with no e ffect on epithelial-cell integrity prevents DU recurrence, we conducte d a randomized double-blind trial of phenoxymethylpenicillin (PEN), 2. 4 twice daily, and placebo (PLA). Patients with an active DU and posit ive H. pylori culture from antral biopsy specimens were treated with 4 0 mg omeprazole daily for 4 weeks, but at week 2 they were allocated a t random to PEN (85 patients) or PLA (85 patients) for up to 14 weeks. Those without recurrence during this treatment were followed up for a nother 6 months. Endoscopy and H. pylori culture were performed at the end of the treatment period and at the end of follow-up, and in betwe en if ulcer symptoms recurred. During the treatment period the ulcer r elapse rate was 5 of 58 (9%) in the PEN group and 34 of 68 (50%) in th e PLA group (P < 0.0001, log-rank test), with 53% and 14%, respectivel y, of the patients in the two groups being H. pylori-negative. The rel apse rate in the PEN group did not differ between H. pylori-negative a nd H. pylori-positive patients. The recurrence rate in the PEN group r emained low for another 5 months but then approached the rate in the P LA group. The prevalence of H. pylori-negative patients at the end of follow-up was 20% in the PEN group and 10% in the PLA group. These dat a provide strong evidence that DU has a bacterial cause, with H. pylor i as the likely agent.