Three-day lansoprazole quadruple therapy for Helicobacter pylori-positive duodenal ulcers: a randomized controlled study

Transcription

1 Aliment Pharmacol Ther 2001; 15: 843±849. Three-day lansoprazole quadruple therapy for Helicobacter pylori-positive duodenal ulcers: a randomized controlled study B. C. Y. WONG*, W. H. WANG*, W.M.WONG*,G.K.K.LAU*,F.M.Y.FUNG*,N.N.S.KUNGà, K. M. CHU, K. C. LAI*, W. H. C. HU*, F. L. HU, X.G.LIU, C.K.CHAN*,M.F.YUEN*,W.M.HUI* &S.K.LAM* Departments of *Medicine and Surgery, University of Hong Kong, Queen Mary Hospital, Hong Kong; Department of Internal Medicine, First Teaching Hospital, Beijing Medical University, PR China and àdepartment of Medicine, United Christian Hospital, Hong Kong Accepted for publication 9 January 2001 SUMMARY Aim: To compare the ef cacy and tolerability of a 3-day quadruple therapy with a standard 7-day triple therapy in eradicating Helicobacter pylori infection and healing duodenal ulcers. Methods: Patients with H. pylori-positive duodenal ulcers were randomized to receive either lansoprazole 30 mg, clarithromycin 500 mg, and metronidazole 400 mg twice daily for 7 days (LCM-7) or lansoprazole 30 mg, clarithromycin 500 mg, metronidazole 400 mg, and bismuth subcitrate 240 mg twice daily for 3 days (LCMB-3). No pre- or post-treatment acid suppression was used. Follow-up endoscopy was performed at week 6. Results: A total of 118 patients were recruited. Sixty patients in the LCM-7 group and 53 patients in the LCMB-3 group returned for endoscopy. Intentionto-treat eradication rates were 87% and 86% (P ˆ 0.94) and per protocol eradication rates were 87% and 94% (P ˆ 0.29) in the LCM-7 and LCMB-3 groups, respectively. Per protocol and intention-to-treat ulcer healing rates were 98% and 98% in LCM-7 and 100% and 91% in LCMB-3, respectively. There were no signi cant differences in ef cacy in relation to the initial metronidazole and clarithromycin susceptibility. Signi- cant reduction in the duration of side-effects was found in the LCMB-3 group. Conclusion: The 3-day quadruple therapy is highly effective, better tolerated and can be considered as a rst-line therapy in duodenal ulcer management. INTRODUCTION Correspondence to: Dr B. C. Y. Wong, Department of Medicine, University of Hong Kong, Queen Mary Hospital, Hong Kong. bcywong@hku.hk Helicobacter pylori eradication has become the standard treatment in duodenal ulcer diseases. 1 Curing of the infection not only resulted in duodenal ulcer healing but signi cantly reduced the risk of ulcer recurrence. 2 One of the preferred regimens according to the Asia Paci c Consensus report was triple therapy which consists of a proton pump inhibitor, clarithromycin and either amoxicillin or metronidazole. 1 Yet, clinicians are still investigating the use of shorter duration treatments, which can reduce the side-effects and improve patient compliance. Others are studying the use of quadruple therapy, aiming at a higher initial success rate and thereby reducing the subsequent induction of resistant strains. We have previously shown that lansoprazole with amoxicillin and metronidazole achieved a per protocol H. pylori eradication rate of 77.5% in some Asian countries, while omeprazole with clarithromycin and metronidazole achieved a better per protocol eradication rate of 95.2% in Hong Kong. 3, 4 We therefore designed Ó 2001 Blackwell Science Ltd 843

2 844 B. C. Y. WONG et al. a new quadruple therapy with shorter duration, hoping to reduce side-effects, improve patient compliance and achieve a high eradication rate. This regimen is compared, in this pilot study, to the best triple therapy regimen in our locality. In addition, we further determined the impact of metronidazole and clarithromycin resistance on the outcome of these regimens. MATERIALS AND METHODS Patients Patients from three endoscopy units in Hong Kong aged between 18 and 80 years with an active duodenal ulcer equal to or greater than 5 mm in size, who were H. pylori-positive (as determined by at least two positive results from CLO test, histology, culture and 13 C-urea breath test) were invited to participate in the study. Patients were excluded if they: had active ulcer bleeding; had undergone previous gastric or duodenal surgery; had underlying malignant conditions; were allergic to the study drugs; had received treatment with bismuth compounds and/or antibiotics and/or had regularly taken NSAIDs in the past 4 weeks; had regularly taken proton pump inhibitors or H 2 -receptor antagonists in the past 2 weeks; or had signi cant concomitant medical disease. The study was approved by the local institutional human research review committee and written informed consent was obtained from all patients before the study. 13 C-urea breath test A 13 C-urea breath test was performed on all patients using 75 mg 13 C-urea and analysed by a purposedesigned mass spectrometer (Fison Instruments, UK). This protocol has been validated in our centre with a sensitivity and speci city of 96.5% and 97.7%, respectively. 5 The results of the breath tests were not disclosed to the investigators, the research nurses or the endoscopists. Bacterial culture and susceptibility testing One antral biopsy was taken during upper endoscopy, and transferred to the laboratory immediately for bacterial culture under microaerophilic conditions at 37 C for between 3 and 7 days. Susceptibility to antibiotics was determined by the E-test. For metronidazole, resistance was de ned by different MIC levels of > 8 mg/l. For clarithromycin, resistance was de ned as an MIC of > 2 mg/l. Treatment Patients were randomized to receive one of the following two regimens: lansoprazole 30 mg plus clarithromycin 500 mg plus metronidazole 400 mg twice daily for 7 days (LCM-7); or lansoprazole 30 mg plus clarithromycin 500 mg plus metronidazole 400 mg plus bismuth subcitrate (DeNol) 240 mg twice daily for 3 days (LCMB-3). Randomization was performed by a drawing a sealed envelope that contained a pre-assigned randomized treatment generated by computer on entry to the study. The treatment was not blinded to the patients, but only blinded to the investigators and the subsequent endoscopists. At all follow-up endoscopies, the endoscopists were blinded to the treatment type and any clinical information related to the patients. No other ulcer healing drugs except antacids were taken before or after the course of treatment until the week-6 follow-up examination. Follow-up Patients were given a symptom diary and returned at week 2, i.e. days 8±14. During this rst visit, the unused tablets were counted, and the symptoms and side-effects were validated with the patients according to the diary card. The total duration (days) of each episode of sideeffects were recorded. They returned again at week 6 for further documentation of symptoms and side-effects, and to repeat endoscopy. The duodenal ulcer was de ned as healed when there was complete epithelialization of ulcer site; partially healed when the size of ulcer was reduced by over 50%; and not healed when the ulcer size was reduced by less than 50%. During endoscopy, one antral biopsy was used for CLO test, one antral and one corpus biopsy were sent for histology, and one antral and one corpus biopsy were sent for culture. 13 C-urea breath tests were also performed to check for H. pylori eradication. Successful eradication was indicated if all these four tests were negative. Statistical analysis The per protocol and intention-to-treat populations were included in the ef cacy analyses. The per protocol population included only those eligible patients who

3 THREE-DAY QUADRUPLE THERAPY FOR H. PYLORI 845 had undergone an upper endoscopy with testing of H. pylori status and had taken at least 75% of the study medication. The intention-to-treat population included all randomized patients who had taken at least one dose of study medication. Patients who had no follow-up were considered not cured. Eradication rates, ulcer healing rates and side-effects event rates of the groups were compared by the v 2 -test with Yates correction and Fisher's exact test. An exact 95% con dence interval (CI) was calculated when appropriate. All P-values below 0.05 were considered to be signi cant. RESULTS Patients population A total of 118 patients with uncomplicated active duodenal ulcer ful lling the inclusion and exclusion criteria were enrolled in this pilot study. The mean age of these patients was 47.9 years (range 18±79 years). Of these, 60 patients were randomized to receive treatment under LCM-7 and 58 patients received treatment under LCMB-3. The baseline characteristics of patients were similar between the two groups (Table 1). These patients' data were all included in the intention-to-treat analysis. Five patients in LCMB-3 group were excluded from the per protocol analysis. Three patients refused the second endoscopy examination, one patient stopped therapy due to side-effects (rash and diarrhoea) and one patient was lost to follow-up. H. pylori eradication and antimicrobial susceptibility According to the intention-to-treat analysis at week 6, the eradication rate of H. pylori was 86.7% and 86.2% Table 1. Demographic data of patients LCM-7 LCMB-3 Total number Age (years) (mean SD) Male : Female 33:27 32:26 Smoking (%) Alcohol (%) Pain score (mean SD) Age onset (years) Duration of symptoms (years) Duodenal ulcer site Anterior (%) Posterior (%) Roof (%) Floor (%) Duodenal ulcer size (mm) (mean SD) in the LCM-7 and LCMB-3 groups, respectively (P ˆ 0.94). For the per protocol analysis, the eradication rate of H. pylori was 86.7% and 94.3% in the LCM- 7 and LCMB-3 groups, respectively (P ˆ 0.29). There was no signi cant difference in the eradication rate between LCM-7 and LCMB-3 (Table 2). For the 118 patients in this study, bacterial culture was performed in 116 patients. The culture positivity rate was 81% (94 out of 116). However, results from susceptibility testing of H. pylori were available in only 83 patients. The remaining 11 strains failed to grow after passage for sensitivity testing. The overall metronidazole (using MIC > 8 mg/l) and clarithromycin resistance rates were 34.6% and 12.3%, respectively. There were no signi cant differences in ef cacy in relation to the initial metronidazole and clarithromycin Table 2. H. pylori eradication rates and ulcer healing rates for each treatment regimen LCM-7 (%) LCMB-3 (%) P-value Intention-to-treat H. pylori eradication (%) 52/60 (86.7) 50/58 (86.2) 0.94 CI: 78.1±95.3 CI: 77.3±95.1 Ulcer healing (%) 59/60 (98.3) 53/58 (91.4) 0.19 CI: 95.0±100 CI: 84.2±98.6 Per protocol H. pylori eradication (%) 52/60 (86.7) 50/53 (94.3) 0.29 CI: 78.1±95.3 CI: 88.1±100 Ulcer healing (%) 59/60 (98.3) 53/53 (100) 0.37 CI: 95.0±100 CI: 100 CI: 95% con dence interval.

4 846 B. C. Y. WONG et al. susceptibility patterns either within the groups or between the groups (Table 3). Using various cut-off MIC levels for metronidazole resistance, there was no signi cant differences in the eradication rates between the sensitive and resistant strains, both in the LCM-7 and LCMB-3 groups, except at an MIC of 128 mg/l in the LCMB-3 group (Table 4). Pre-treatment susceptibility testing was available for eight of the 11 patients who failed eradication. The MIC levels for metronidazole were: 2 mg/l (three strains), 32 mg/l (one strain), 64 mg/l (two strains), and 128 mg/l (two strains). Post-treatment susceptibility testing was available in four patients; all had double resistance to metronidazole and clarithromycin. The MIC levels for metronidazole were: 32 mg/l (two strains), 64 mg/l (one strain), 256 mg/l (one strain). Ulcer healing The ulcer healing rates (intention-to-treat) were 98.3% and 91.4% for LCM-7 group and LCMB-3 group, respectively. There was no signi cant difference between the two groups, both in intention-to-treat and per protocol analysis (Table 2). Side-effects The total numbers of episodes of side-effects were 85 and 65 in the LCM-7 and LCMB-3 groups, respectively (Table 5). The total number of patients experiencing side-effects were 47 out of 60 (78.3%) and 44 out of 56 (78.6%) in the LCM-7 and LCMB-3 groups, respectively. The most common side-effects were bitter taste (51.7%), altered stool frequency/consistency (24.1%), dizziness (11.2%), malaise (11.2%) and dark stool (9.5%) (Table 5). Although one patient in the LCMB-3 group terminated the study as a result of adverse events (rash and diarrhoea), no severe side-effects occurred in the two groups. There was no signi cant difference in the number of patients having side-effects between the two groups (P ˆ 0.59). However, for each episode of sideeffects, the LCMB-3 group had signi cantly shorter duration compared with the LCM-7 group (mean duration: 2.5 days vs. 5.5 days, P < 0.001). For each patient, the LCMB-3 group also had signi cantly shorter duration of side-effects compared with the LCM-7 group (mean duration: 2.6 days vs. 6.2 days, P < 0.001). A cohort of 60 patients with duodenal ulcer will suffer an additional days of side-effects for using the LCM-7 rather than the LCMB-3 regimen. LCM-7 (%) LCMB-3 (%) Metronidazole* Susceptible 22/24 (91.7) CI: 80.7±100 28/29 (96.6) CI 90.0±100 Resistant 14/17 (82.4) CI: 61.9±98.1 9/11 (81.8) CI: 60.5±100 P-value Clarithromycin Susceptible 32/35 (91.4) CI: 82.1±100 34/36 (94.4) CI: 86.9±100 Resistant 4/6 (66.7) CI: 29.0±100 3/4 (75.0) CI: 32.6±100 Table 3. H. pylori eradication rates according to pre-treatment antimicrobial susceptibility * Metronidazole resistance was de ned as MIC > 8 mg/l in this table. P-value is based on comparison between the metronidazole-susceptible and -resistant strains within each regimen. Table 4. H. pylori eradication rates according to different cut-off values for metronidazole resistance LCM-7 Eradication rate (%) LCMB-3 Eradication rate (%) Total Eradication rate (%) MIC (mg/l) S R P-value S R P-value S R P-value 8 22/24 (91.7) 14/17 (82.4) /29 (96.6) 9/11 (81.8) /53 (94.3) 23/28 (82.1) /26 (92.3) 12/15 (80) /30 (96.7) 8/10 (80) /56 (94.6) 20/25 (80) /27 (88.9) 12/14 (85.7) /30 (96.7) 8/10 (80) /57 (93.0) 20/24 (83.3) /37 (86.5) 4/4 (100) /34 (97) 4/6 (66.7) /71 (91.5) 8/10 (80) /39 (87.2) 2/2 (100) /35 (97.1) 3/5 (60) /74 (91.9) 5/7 (71.4) 0.28 S, metronidazole-susceptible strains; R, metronidazole-resistant strains.

5 THREE-DAY QUADRUPLE THERAPY FOR H. PYLORI 847 Table 5. Side effects experienced by patients LCM-7 (%) LCMB-3 (%) Side effects (no. of patient ˆ 60) Mean days (no. of patient ˆ 56) Mean days Bitter taste 34 (57) (46) 2.7 Bowel disturbance 21 (35) (8) 2.6 Dizziness 9 (15) (7) 2.3 Malaise 7 (12) (10) 1.7 Dark stool 1 (2) (17) 2.7 Others 13 (22) (18) 2.6 Total episodes * * Total patients (%) 47 (78) 6.15* 44 (79) 2.59* * P < for total episodes and total patients comparing between LCM-7 and LCMB-3 groups. Accuracy of 13 C-urea breath test We evaluated the accuracy of the 13 C-urea breath test according to a gold standard (histology + CLO test and/ or culture). All 118 patients underwent a urea breath test before treatment. The sensitivity and speci city was 98.3% and 100%, respectively. A post-treatment urea breath test was performed in 112 patients, with sensitivity and speci city of 90.9% and 100%, respectively. Overall, the sensitivity was 97.7% and the speci city was 100%. DISCUSSION In this pilot study, we compared the ef cacy and tolerability of a 3-day lansoprazole quadruple therapy with those of a 7-day lansoprazole triple therapy. Two regimens tested in this study were found to be highly effective, with per protocol eradication rates of 94.3% in the 3-day regimen and 86.7% in the 7-day regimen. The reported eradication rates of lansoprazole, clarithromycin and metronidazole (LCM) triple therapy have varied from 78% to 92.3%. 6±9 Few studies have focused on the impact of pre-treatment antimicrobial resistance on the outcome of this regimen. 7 Harris et al. reported that H. pylori was eradicated in 92% of patients with metronidazole-susceptible strains compared with 75% of patients with metronidazole-resistant strains. 7 However, in our study, no signi cant difference was observed in the eradication rates between metronidazole-susceptible and -resistant strains, possibly due to a small number in each group. So far there has been no report on the impact of clarithromycin resistance on the ef cacy of LCM therapy. The eradication rate of clarithromycin-susceptible strains appears to be higher than clarithromycin-resistant strains in our pilot study, but the patient number is too small for a valid comparison. Further studies are warranted to explore the use of lansoprazole-based triple therapy in clarithromycin-resistant H. pylori infection. In other studies, high eradication rates were achieved even for metronidazole- or clarithromycin-resistant strains. 10 This may be explained by synergy between a proton pump inhibitor and clarithromycin, or possibly the potency of clarithromycin as an anti-h. pylori agent may in itself overcome any in vitro metronidazole resistance. 11±13 More studies need to be performed. One week quadruple therapy of proton pump inhibitor + metronidazole + tetracycline + bismuth subcitrate was frequently studied, with the H. pylori eradication rate ranging from 86% to 100%. 14±18 The eradication rate was similar if the same treatment was given for 4 days, but poor if the duration was further 14, 19±23 reduced to 2 days. We therefore designed a 3-day quadruple therapy substituting tetracycline with clarithromycin, with no pre- or post-treatment acid suppression. The short regimen was still able to achieve a high per protocol eradication rate of 94.3%. This is in contrast to a previous study of 3-day quadruple therapy with lansoprazole + clarithromycin + amoxicillin + bismuth, showing a per protocol eradication rate of 25%, 24 but in agreement with another 3-day quadruple therapy of octreotide + amoxicillin + metronidazole + bismuth, showing an eradication rate of 88.5%. 25 We have focused on documenting the side-effects, which may account for the relatively high incidence compared with other studies. 6, 8, 10, 26, 27 However, patient compliance was very good in this study and only one patient withdrew from the study because of side-effects. One of the major ndings from our study was the signi cant reduction in the number of days with side-effects with the use of a shorter regimen. By shortening the duration of

6 848 B. C. Y. WONG et al. treatment by 4 days, the number of days with side-effects was reduced by an average of 3.5 days per patient. This was very important in considering the cost savings for patients and society. We therefore advocate the use of this 3-day quadruple therapy. We conclude that 3-day H. pylori eradication therapy with lansoprazole 30 mg, clarithromycin 500 mg, metronidazole 400 mg and bismuth subcitrate 240 mg twice daily is highly effective, well-tolerated, and associated with fewer side-effects. This new regimen should be studied further with a larger sample size to see whether it can be used as the rst line treatment for H. pylori-positive duodenal ulcer diseases. ACKNOWLEDGEMENTS We thank our endoscopy nurses in various centres. We are grateful to Dr Jia-Qing Huang for statistical advice. This study was partly supported by the Peptic Ulcer Research Fund, University of Hong Kong and the Simon KY Lee Gastroenterology Research Fund. REFERENCES 1 Lam SK, Talley NJ. Report of the 1997 Asia Paci c Consensus Conference on the management of Helicobacter pylori infection. J Gastroenterol Hepatol 1998; 13: 1±12. 2 Wong BCY, Lam SK, Lai KC, et al. Triple therapy for Helicobacter pylori eradication is more effective than long-term maintenance anti-secretory treatment in the prevention of recurrence of duodenal ulcer: a prospective long-term followup study. Aliment Pharmacol Ther 1999; 13: 303±9. 3 Wong BCY, Xiao SD, Hu FL, et al. Comparison of lansoprazolebased triple and dual therapy for treatment of Helicobacter pylori-related duodenal ulcer: An Asian multi-center prospective double blind randomized placebo controlled study. Aliment Pharmacol Ther 2000; 14: 217±24. 4 Chu KM, Choi HK, Tuen HH, Law SYK, Branicki FJ, Wong J. A prospective randomized trial comparing the use of omeprazole-based dual and triple therapy for eradication of Helicobacter pylori. Am J Gastroenterol 1998; 93: 1436±42. 5 Wong WM, Wong BCY, Wong KW, et al. 13 C-urea breath test without a test meal is highly accurate for the detection of Helicobacter pylori infection in Chinese. Aliment Pharmacol Ther 2000; 14: 1353±8. 6 Perng CL, Kim JG, El-Zimaity HM, Osato MS, Graham DY. Oneweek triple therapy with lansoprazole, clarithromycin, and metronidazole to cure Helicobacter pylori infection in peptic ulcer disease in Korea. Dig Dis Sci 1998; 43: 464±7. 7 Harris AW, Pryce DI, Gabe SM, et al. Lansoprazole, clarithromycin and metronidazole for seven days in Helicobacter pylori infection. Aliment Pharmacol Ther 1996; 10: 1005±8. 8 Takimoto T, Satoh K, Taniguchi Y, et al. The ef cacy and safety of one-week triple therapy with lansoprazole, and metronidazole for the treatment of Helicobacter pylori infection in Japanese patients. Helicobacter 1997; 2: 86±91. 9 Schwartz H, Krause R, Siepman N, et al. Seven-day triple therapy with lansoprazole, clarithromycin, and metronidazole for the cure of Helicobacter pylori infection: a short report. Helicobacter 1996; 1: 251±5. 10 Misiewicz JJ, Harris AW, Bardhan KD, et al. Lansoprazole Helicobacter Study Group. One week triple therapy for Helicobacter pylori: a multicentre comparative study. Gut 1997; 41: 735±9. 11 van Caekenberghe DI, Breyssend J. In vitro synergistic activity between bismuth subcitrate and various antimicrobial agents against Campylobacter pyloridis (C. pylori). Antimicrob Agents Chemother 1987; 31: 1429± Gustavsson LE, Kaiser JF, Edmonds AL. Effect of omeprazole on concentrations of clarithromycin in plasma and gastric tissue at steady state. Antimicrob Agents Chemother 1995; 39: 2078± Peterson WL, Graham DY, Marshall B. Clarithromycin as mono-therapy for eradication of H. pylori: a randomized, double-blind trial. Am J Gastroenterol 1993; 88: 1860±4. 14 Kung NN, Sung JJ, Yuen NW, et al. Anti-Helicobacter pylori treatment in bleeding ulcers: randomized controlled trial comparing 2-day versus 7-day bismuth quadruple therapy. Am J Gastroenterol 1997; 92: 438± De Boer WA, Van Etten RJ, van de Wouw BA, Schneeberger PM, van Oijen AH, Jansen JB. Bismuth-based quadruple therapy for Helicobacter pyloriða single triple capsule plus lansoprazole. Aliment Pharmacol Ther 2000; 14: 85±9. 16 Gomollon F, Ducons JA, Ferrero M, et al. Quadruple therapy is effective for eradicating Helicobacter pylori after failure of triple proton pump inhibitor-based therapy: a detailed, prospective analysis of 21 consecutive cases. Helicobacter 1999; 4: 222±5. 17 Korman MG, Bolon TD, Nicholson FB, Engelman JL. Lansoprazole quadruple therapy is effective in curing Helicobacter pylori infection. Helicobacter 1998; 3: 202±5. 18 Gomollon F, Ducons JA, Gimeno L, et al. The ideal therapy must be de ned in each geographical area: experience with a quadruple therapy in Spain. Helicobacter 1998; 3: 110±4. 19 de Boer WA, van Etten RJ, Schade RW, Ouwehand ME, Schneeberger PM, Tytgat GN. 4-day lansoprazole therapy: a highly effective cure for Helicobacter pylori infection. Am J Gastroenterol 1996; 91: 1778± de Boer WA, van Etten RJ, Schneeerger PM. Four-day lansoprazole-quadruple therapy in the routine treatment of Helicobacter pylori infection. Neth J Med 1998; 52: 10±5. 21 de Boer SY, Siem TH. Four-day quadruple therapy as a routine treatment for Helicobacter pylori infection. Aliment Pharmacol Ther 1997; 11: 1119± de Boer WA, van Ette RJ, Coremans A, Schneeberger PM. Two-day `weekend' lansoprazole-quadruple therapy for Helicobacter pylori infection. Aliment Pharmacol Ther 1998; 12: 77± Calvet X, Garcia N, Campo R, Brullet E, Comet R, Navarro M. Two-day quadruple therapy for cure of Helicobacter pylori

7 THREE-DAY QUADRUPLE THERAPY FOR H. PYLORI 849 infection: a comparative, randomized trial. Am J Gastroenterol 1998; 93: 932±4. 24 Grimley CE, Penny A, O'sullivan M, et al. Comparison of two 3-day Helicobacter pylori eradication regimens with a standard 1-week regimen. Aliment Pharmacol Ther 1999; 13: 869± Ladas SD, Malamou-Lada H, Economou G, Tassios PS, Raptis SA. A three-day octreotide-containing Helicobacter pylori eradication therapy for cure of peptic ulcers. Hepatogastroenterology 1998; 45: 761± Labenz J, Stolte M, Peitz U. One-week triple therapy with omeprazole, amoxycillin, and either clarithromycin or metronidazole for cure of Helicobacter pylori infection. Aliment Pharmacol Ther 1996; 10: 207± Lind T, van Zanten SV, Unge P, et al. Eradication of Helicobacter pylori using one-week triple therapies combined omeprazole with two antimicrobials: the MACH I study. Helicobacter 1996; 1: 138±44.