We compared the efficacy of three dual and two triple therapies for eradication of Helicobacter pylori (HP), and evaluated the influence of smoking and omeprazole pretreatment on HP eradication. 220 patients with proven HP infection (histology and 13C-urea breath test [UBT]) were randomly allocated to one of the following regimes: BMT (bismuth subsalicylate 600 mg t. i. d. for 28 days, metronidazole 400 mg t. i. d. and tetracycline 500 mg q. i. d. for ten days). OA (omeprazole 40 mg o. d. and amoxicillin 750 mq q. i. d. for 14 days), OC (omeprazole 40 mg o. d. and clarithromycin 500 mg b. i. d. for 14 days), OT (omeprazole 40 mg o. d. and tetracycline 500 mg q. i. d. for 14 days), OMC (omeprazole 40 mg o. d., metroinidazole 400 mg t. i. d. and clarithromycin 250 mg b. i. d. for seven days). Eradication was defined as negative UBT six weeks after completion of the therapy. In an all-patients-treated (per-protocol) analysis, the eradication rates were: BMT, 91% (93%); OA, 84% (90%); OC, 74% (74%); OT, 24% (24%); and OMC, 90% (93%). Smoking impaired the success of OA and OT (p < 0.05), but the efficacy of the triple regimens was not affected. Omeprazole pretreatment did not influence eradication rates. Thus, highest eradication rates were achieved with the two triple therapies tested. However, OA, given at a daily antibiotic dose of 3 g amoxicillin for 14 d, was also highly effective. After failure of triple therapy, OA was successful in seven of ten patients (70%). The efficacy of OC was lower than that of the triple therapies (p < 0.05). In conclusion, metronidazole- and clarithromycin-based triple therapies are highly effective first line therapies. OA, given at a dose of 3 g per day over 14 days, should be considered as a possible second line therapy, e.g. in retherapy after failed triple therapy.