Volume 4 • Number 2 • 1999 HELICOBACTER
Pretreatment Antibiotic Resistance in Helicobacter pylori Infection: Results of Three Randomized Controlled Studies Giuseppe Realdi,* Maria P. Dore,*§ Andrea Piana,† Antonella Atzei,* Monica Carta,* Luigi Cugia,* Alessandra Manca,‡ Bianca M. Are,† Giovanni Massarelli,‡ Ida Mura,† Alessandro Maida,† and David Y. Graham§ *Departments of Internal Medicine, †Institute of Hygiene and Preventive Medicine and ‡Institute of Histopathology, University of Sassari, Italy; §VA Medical Center and Baylor College of Medicine, Houston, TX, USA
ABSTRACT
Background. Although combinations of antibiotics and antisecretory drugs are useful for treatment of Helicobacter pylori infection, treatment failure is common. The aim of this study was to evaluate the relation between pretreatment antibiotic resistance and outcome by using six different treatment regimens for H. pylori infection. Patients and Methods. Three hundred sixty-nine consecutive H. pylori–infected patients with dyspeptic symptoms were enrolled in three consecutive randomized, controlled, single-center clinical trials: trial A, 128 patients; trial B, 125 patients; trial C, 116 patients. Treatments consisted of (A) a 15-day course of dual therapy (omeprazole, 20 mg bid, and amoxicillin, 1 gm bid, or clarithromycin, 500 mg tid) (OA vs OC); (B) a 7-day triple therapy of omeprazole, 20 mg bid, plus metronidazole, 500 mg bid, and amoxicillin, 1,000 mg bid, or clarithromycin, 500 mg tid (OMA vs OMC); or (C) omeprazole, 20 mg bid, plus metronidazole, 500 mg bid, plus tetracycline, 500 mg qid, or doxycycline, 100 mg tid (OMT vs OMD). Diagnostic endoscopy was made in all patients before and 5 to 6 weeks after therapy. Six biopsies were taken from each patient for histology, rapid urease test, and H. pylori culture; antibi-
otic susceptibility testing was performed using the E-test method. Results. Overall cure rates were poor for both dual therapies OA and OC (38% and 37%, respectively) and for triple therapies OMA, OMC, and OMD (57%, 55%, and 58%, respectively). The OMT combination was successful in 91% (95% confidence interval [CI], 80.4%–97%). Metronidazole resistance was present in 29.7% (95% CI, 24%–35%), amoxicillin resistance was present in 26% (95% CI, 21%–32%), clarithromycin resistance was present in 23.1% (95% CI, 18%– 29%), tetracycline resistance was present in 14% (95% CI, 10%–20%), and doxycycline resistance was present in 33.3% (95% CI, 21%–47%). Antibiotic resistance markedly reduced the cure rates and accounted for most of the poor results with the triple therapies: 89% versus 23%; 77% versus 26%; 100% versus 60%; and 67% versus 23% for OMC, OMA, OMT, and OMD, respectively. OMT appeared to be the best because of the high success rate with metronidazole-resistant H. pylori (71%) and in low-level tetracycline resistance. Conclusions. Pretreatment antibiotic-resistant H. pylori can, in part, explain the low cure rate of the infection and the variability in outcome in reported trials.
T
clarithromycin, tetracycline, or metronidazole) administered for 1 to 2 weeks [2–4]. The efficacy of anti–H. pylori treatment, however, is increasingly being undermined by antibiotic resistance of H. pylori, especially to metronidazole and clarithromycin [5]. Because of poor results previously obtained in clinical practice in our unit using traditional regimens for H. pylori infection, we initiated studies employing different antimicrobial regimens to compare efficacy to pretreatment antibiotic resistance. For this reason, three consecutive, prospective, randomized trials were conducted enrolling consecutive dyspeptic patients who tested positively for H. pylori infection.
he 1994 National Institute of Health Consensus Development Conference recommended anti– Helicobacter pylori therapy for every patient with peptic ulcer [1]. Since that time, a variety of treatment protocols have been approved by regulatory agencies in different countries. Cure rates in the range of 80% to 95% usually are expected with triple therapies consisting of bismuth or a proton pump inhibitor and two antibiotics (amoxicillin,
Reprint requests to: Prof. Giuseppe Realdi, Istituto di Clinica Medica Generale e Terapia Medica, Viale San Pietro, 8, 07100 Sassari, Italy. E-mail:
[email protected] © Blackwell Science, Inc. 1083-4389/99/$10.50/106 106–112
106
107
Antibiotic Resistance and H. pylori Eradication Materials and Methods
Patient Selection Between July 1995 and July 1997, consecutive H. pylori–infected patients with dyspeptic symptoms referred to the Digestive Endoscopy Unit, Department of Internal Medicine, University of Sassari, Italy, were allocated randomly to one of two treatment options in three consecutive single-center, prospective, controlled clinical trials. Dyspeptic symptoms were defined as at least 3 months’ experience of epigastric pain, upper abdominal discomfort, nausea and vomiting, early satiety, heartburn, and regurgitation. Each symptom was scored from 0 to 3 with regard to frequency and intensity and was assessed by a structured questionnaire. Randomization was done by use of sealed envelopes containing one of the therapies prepared from a list of random numbers generated by a computer. After the regimen was identified on opening the sealed envelope, each enrolled patient took medication as an ordinary prescription. No pharmaceutical company participation was involved.
Exclusions Patients who received bismuth compounds, antisecretory drugs, or antibiotics during the 4 weeks before endoscopy were excluded. Other exclusion criteria included gastrointestinal surgery; pregnancy or lactation; regular use of nonsteroidal antiinflammatory drugs, including acetylsalicylic acid; malignancy; known allergy to penicillin; severe liver, heart, or kidney disease; or prior treatment for H. pylori infection. Written informed consent was obtained from all patients, and the protocol was approved by the ethics committee of the faculty of medicine, University of Sassari.
Diagnostic Methods In each patient, four biopsy specimens were taken from the antrum, two from the angulus, and two from the corpus of the stomach. Biopsy forceps were sterilized by autoclaving, and endoscopes were disinfected using an automatic washing machine (Olympus ETD2; Olympus Corp., Lorenzatto MG, S.p.A., Torino, Italy). Biopsy specimens for histology were fixed immediately in 10% buffered formalin. Two biopsies from the antrum, two from the angulus, and two from the corpus were stained with hematoxylin-eosin (H&E) and Giemsa stains to grade the density of H. pylori. Gastritis was assessed using the Sydney system [6] by a pathologist who was unaware of the clinical data.
One biopsy from the antrum was placed immediately into a 6% urea solution for a rapid urease test (CP-test; Yamanouchi S.p.A., Milan, Italy). A positive result was recorded for a color change from yellow to pink within 24 hours. H. pylori infection was defined by the contemporary detection of a positive rapid urease test and the presence of H. pylori on histological examination of gastric biopsies.
Microbiology One antral biopsy from each patient was transported immediately at room temperature in Portagerm pylori (BioMerieux, S.p.A., Rome, Italy) for culture. Biopsies were streaked on Columbia agar plates (BioMerieux) and were incubated for 6 to 8 days at 378C, 12% CO2 (Campy Pak Plus, BBL Becton Dickinson, Cockesville, MD), and 100% relative humidity. Bacterial growth was identified as H. pylori on the basis of colony morphology and positive biochemical reactions for catalase, urease, oxidase, gamma glutamyl transferase, alkaline phosphatase, and negative hippurate hydrolysis and nitrate reduction tests. The pretreatment minimal inhibitory concentration (MIC) for amoxicillin, clarithromycin, metronidazole, doxycycline, and tetracycline were determined by the Epsilometer test (AB Biodisk, Uppsala, Sweden), following the manufacturer’s instructions, and were read after more than 3 days of incubation. All tests were made in duplicate. The breakpoints used to define resistant H. pylori were as follows: amoxicillin resistance, MIC greater than 8 mg/ml; clarithromycin resistance, MIC greater than 2 mg/ml; metronidazole resistance, MIC greater than 8 mg/ml; doxycycline resistance, MIC greater than 2 mg/ml; and tetracycline resistance, MIC greater than 2 mg/ml [7]. Posttreatment culture of biopsy specimens and antibiotic susceptibility tests were not performed.
Treatment Regimens H. pylori–positive patients were enrolled in six different therapeutic regimen groups in three prospective, consecutive, and sequential randomized trials (Tables 1–3). Trial A consisted of omeprazole plus amoxicillin versus omeprazole plus clarithromycin (OA vs OC). Treatment was 15 days of therapy with omeprazole (20 mg bid) plus amoxicillin (1,000 mg bid) or plus clarithromycin (500 mg tid). Trial B consisted of omeprazole plus amoxicillin plus metronidazole versus omeprazole plus clarithromycin plus metronidazole (OMA vs OMC). Treatment was 7 days with omeprazole (20 mg bid)
108
Realdi et al.
plus amoxicillin (1,000 mg bid) and metronidazole (500 mg bid) or clarithromycin (500 mg tid) and metronidazole (500 mg bid). Trial C consisted of omeprazole plus tetracycline plus metronidazole versus omeprazole plus doxycycline plus metronidazole (OMT vs OMD). Treatment was 7 days with omeprazole (20 mg bid) plus tetracycline (500 mg qid) and metronidazole (500 mg bid) or doxycycline (100 mg tid) and metronidazole (50 mg bid).
Outcome Assessment The primary end point was cure of the H. pylori infection. Success of treatment was defined as absence of H. pylori organisms on histological examination of gastric biopsies obtained from the antrum, angulus, and gastric corpus 40 to 50 days after therapy was completed.
Compliance and Side Effects To enhance compliance, verbal and printed instructions were given to the patients, and a telephone service was available for questions. Patients were asked to count all tablets that they took on any given day and to return the medicine boxes on the next visit. After completing treatment, patients were evaluated for compliance or possible side effects at the first endoscopy follow-up by a physician, both by being questioned and by pill count. Compliance was scored according to the number of days the study medication was taken. Compliance was considered good if at least 90% of the total number of the pills were taken. Side effects were graded as mild (did not limit daily activities), moderate (limited daily activities to some extent), or severe (made daily activities all but impossible).
Statistics The goal was to enter approximately 60 patients per group to have a sample size sufficient to result in 95% confidence intervals (CIs) that would pro-
Received standard intervention as allocated Followed up Withdrawn Intervention ineffective Lost to follow-up Completed trial
OMA
OMC
63 63 0 27 0 63
62 60 2 27 0 60
OMA, omeprazole plus metronidazole plus amoxicillin; OMC, omeprazole plus metronidazole plus clarithromycin. Note: Registered or eligible patients number 200. Of these, 75 were not randomized for the following reasons: 71 patients were Helicobacter pylori–negative, 3 retained food, and 1 refused endoscopy.
vide a good estimate of the efficacy of any effective therapy [8,9]. Categorical data were analyzed by Fisher’s exact test or the chi-squared test whenever it was appropriate using SigmaStat software (SPSS Inc, Chicago, IL). Ninety-five percent CIs were calculated using a program written by Fagan [10].
Results A total of 369 patients participated in these trials, of whom 128 patients were included in trial A, 125 in trial B, and 116 in trial C. The patient baseline characteristics and the results for the three clinical trails are summarized in Tables 4 and 5. No statistically significant differences were found in cure rate with any treatment with regard to age, gender, smoking, alcohol consumption, endoscopic findings, or compliance.
Trial A Trial A consisted of administration of omeprazole plus amoxicillin versus omeprazole plus clarithromycin (OA vs OC). Sixty-five patients received OA, and 63 patients received OC. One patient dropped out of the OC group, and two dropped out of the OA group. H. pylori therapy succeeded (per protocol) in 38% of evaluable OA patients
Table 3 Flow chart of patients enrolled in trial C
Table 1 Flow chart of patients enrolled in trial A
Received standard intervention as allocated Followed up Withdrawn Intervention ineffective Lost to follow-up Completed trial
Table 2 Flow chart of patients enrolled in trial B
OA
OC
65 63 2 39 0 63
63 62 1 39 0 62
OA, omeprazole plus amoxicillin; OC, omeprazole plus clarithromycin. Note: Registered or eligible patients numbered 206. Of these, 78 were not randomized for the following reasons: 75 patients were Helicobacter pylori–negative, 2 refused endoscopy and, in 1, varices were present.
Received standard intervention as allocated Followed up Withdrawn Intervention ineffective Lost to follow-up Completed trial
OMT
OMD
56 56 0 5 0 56
60 57 3 24 0 57
OMA, omeprazole plus metronidazole plus tetracycline; OMC, omeprazole plus metronidazole plus doxycyclin. Note: Registered or eligible patients numbered 186. Of these, 70 were not randomized for the following reasons: 68 patients were Helicobacter pylori– negative and 2 retained food.
109
Antibiotic Resistance and H. pylori Eradication (24 of 63) as compared to 37% in the OC group (23 of 62; p 5 1.0; see Tables 1 and 5).
Table 5 Results of the three clinical trials Trial A OA
Trial B Trial B consisted of administration of omeprazole plus amoxicillin plus metronidazole (OMA) versus omeprazole plus clarithromycin plus metronidazole (OMC). Sixty-three patients received OMA, and 62 received OMC. H. pylori treatment was successful in 57% (36 of 63) of OMA patients and in 55% (33 of 60) of OMC patients (p 5 .85) (see Tables 2 and 5).
OC
Trial B AMA
OMC
Trial C OMT
OMD
Side effects (%) 6 13 8 15 16 5 Compliance , 90% 0 1 0 2 0 0 Drop-out 2 1 0 2 0 3 Cure rate % PP 38 37 57 55 91 58 95% CI (%) 21–51 25–50 44–70 42–68 80–97 44–71 % ITT 95% CI (%)
37 36 57 53 91 55 25–50 25–49 44–70 40–66 80–97 42–68
O, omeprazole; A, amoxicillin; C, clarithromycin; M, metronidazole; T, tetracycline; D, doxycycline; PP, per protocol; ITT, intention to treat; CI, confidence interval.
Trial C Trial C consisted of administration of omeprazole plus tetracycline plus metronidazole (OMT) versus omeprazole plus doxycycline plus metronidazole (OMD). Fifty-six patients received OMT, and 60 received OMD. Three OMD patients refused to undergo the second endoscopy and were omitted from the per-protocol analysis. The cure rate was 91% (51 of 56) for OMT and was significantly greater than the 57.8% cure rate (33 of 57) for treatment with OMD (p 5 .0001) (see Tables 3 and 5).
H. pylori Culture and Effect of Antibiotic Resistance Cultures of pretreatment biopsy specimens were positive in 80.4% of the patients (297 of 369) enrolled in the three studies. Susceptibility testing was not possible in 72 because of fungal overgrowth, premature death of the isolate, or absence of microbial growth. Pretreatment susceptibility testing showed amoxicillin resistance in 26% (61 of 232); clarithromycin resistance in 23.1% (51 of 220); metronidazole resistance in 29.7% (83 of 279); tetracycline resistance in 14% (29 of 206); and doxycycline resistance in 33.3% (18 of 54). The susceptibility to amoxicillin was also tested
Table 4 Patient baseline characteristics for the three clinical trials Trial A
No. of patients Men:women Mean age (yr) Smokers (%) Peptic ulcer (%)
Trial B
Trial C
OA
OC
OMA
OMC
OMT
OMD
65 29:33 53 31 9
63 36:26 52 19 8
63 34:29 48 21 8
62 50:16 50 29 6
56 35:21 50 34 9
60 35:25 51 25 5
O, omeprazole; A, amoxicillin; C, clarithromycin; M, metronidazole; T, tetracycline; D, doxycycline.
using the agar dilution method, and similar levels of resistance were determined (unpublished data). Besides susceptibility testing for the specific drug used in each subgroup, we obtained contemporary pretreatment susceptibility tests for amoxicillin, clarithromycin, metronidazole, and tetracycline in 206 patients. Thirty-one patients (15%) harbored H. pylori resistant to amoxicillin and clarithromycin, and 19 H. pylori isolates (9%) were resistant to amoxicillin, clarithromycin, and metronidazole. With all therapy regimens, H. pylori resistance resulted in marked reduction in cure rate, although the difference was statistically significant only in triple-therapy regimens (Table 6). In four patients, H. pylori was successfully eradicated despite resistance to both drugs.
Side Effects OA Versus OC. All OA patients (65) and 98% of OC patients (62 of 63) completed the 15-day course of therapy. One OC patient stopped treatment after 9 days because of side effects. Two patients in the amoxicillin group refused the followup endoscopy (see Table 1). No side effects were reported in 94% of the OA group (61 of 65); 4 patients experienced diarrhea, nausea, and abdominal pain. In the OC group, five patients reported abdominal pain, bloating, and diarrhea, and three had the same symptoms plus headache. Side effects were mild, and only one patient stopped treatment. OMA Versus OMC. Overall, the OMA and OMC treatment regimens were well tolerated; two patients were unable to complete the 1-week triple therapy in the OMC group. These two patients stopped treatment at the fourth and fifth days, respectively, because of severe bloating, abdominal pain, and diarrhea (see Table 2). Nine patients in the OMC group experienced, headache, abdomi-
110
Realdi et al.
Table 6 Pretreatment antimicrobial sensitivity and effect of resistance on therapy outcome Regimen
No. of patients with positive culture
OA
47
OC
49
OMA
49
OMC
50
OMT
48
OMD
54
Susceptibility
No. of patients
Cured (%)
95% CI
p Value
Sensitive to amoxicillin Resistant to amoxicillin Sensitive to clarithromycin Resistant to clarithromycin Sensitive to both Resistant to either Resistant to metronidazole alone Resistant to amoxicillin alone Resistant to both Sensitive to both Resistant to either Resistant to metronidazole alone Resistant to clarithromycin alone Resistant to both Sensitive to both Resistant to either Resistant to metronidazole alone Resistant to tetracycline alone Resistant to both Sensitive to both Resistant to either Resistant to metronidazole alone Resistant to doxycycline alone Resistant to both
35 12 36 13 22 27 14 10 3 28 22 12 6 4 34 15 7 3 5 24 30 12 14 4
19 (54%) 3 (25%) 15 (42%) 2 (15%) 17 (77%) 7 (26%) 4 (28%) 2 (20%) 1 (33%) 25 (89%) 5 (23%) 3 (25%) 2 (33%) 0 34 (100%) 9 (60%) 5 (71%) 1 (33%) 3 (60%) 16 (67%) 7 (23%) 5 (42%) 2 (14%) 0
37–71
.10
26–59
.10
55–92
.0005
72–98
.0001
90–100
.0001
45–84
.002
CI, confidence interval; O, omeprazole; A, amoxicillin; C, clarithromycin; M, metronidazole; T, tetracycline; D, doxycycline.
nal pain, and bloating. One also reported metallic taste, and two experienced vomiting and nausea. Two women developed vaginal candidiasis. Overall, side effects were mild and did not interfere with the daily activities of the patients. Five patients in the OMA group developed mild side effects (abdominal pain and bloating).
OMT Versus OMD. Three patients withdrew from the OMD group because of significant side effects: One experienced severe heartburn; the other two had nausea and abdominal bloating (see Table 3). The adverse events most often encountered in the OMT group were sore mouth and tongue (n 5 4), nausea (n 5 8), easy fatigability (n 5 6), and vaginal candidiasis (n 5 5). All side effects were mild, and all patients completed therapy. Discussion Although H. pylori infection is susceptible to many different antibiotics, successful treatment remains a challenge. Current research is focused on identifying treatment regimens with high efficacy in different populations and an acceptable low rate of side effects. One problem is that there is considerable variability in the outcome of studies that purportedly use the same combination of drugs [2–4,11,12]. Although reasons for such differences are unclear, problems with compliance or with drug-resistant H.
pylori are thought to be the most important factors [12,13]. Overall, we obtained poor results with the widely used triple-therapy combinations of omeprazole and metronidazole plus clarithromycin or amoxicillin. Antimicrobial resistance led to a marked reduction in effectiveness in every combination studied, as cure rates were dependent on pretreatment antibiotic sensitivity (see Table 6). For example, we obtained a poor cure rate (55%) with clarithromycin-based triple therapy (OMC), a regimen in which pooled results from 21 treatment arms reported a cure rate of 88.7%, (95% CI 5 86.8%–90.5%) [14]. The cure rate for OMC was similar to that expected (89%) with sensitive H. pylori but fell to 23% with antibiotic-resistant H. pylori. The high prevalence of antibiotic resistance in H. pylori in our region (Northern Sardinia) highlights the detrimental effect of antimicrobial resistance on outcome of H. pylori therapy. Specifically, we obtained a particularly high prevalence of resistance to amoxicillin and to tetracycline (26% and 14%, respectively) as compared to previous reports, in which resistance to these drugs was found to be very rare. Pharmacoepidemiological data about antibiotic resistance to other microorganisms are not available in our region, but we cannot exclude the possibility that antibiotic resistance could be a more diffuse phenomenon owing
111
Antibiotic Resistance and H. pylori Eradication to the widespread use of antimicrobial drugs, especially in the pediatric age groups. We have recently described the penicillin tolerance phenomenon in H. pylori [15,16]. It appears likely that one of the major problems with understanding the variability in outcome of reported studies relates to the failure to report separately the outcome in sensitive and resistant strains. In many treatment trials of H. pylori infection, the prevalence of resistance is not reported and, when noninvasive tests such as the urea breath test are used as the primary measure of the presence or absence of infection, resistance is not discernible [17–19]. In the absence of data regarding resistance patterns, only OMT appeared to be a useful therapy in North Sardinia. This apparent superiority was based on the relatively high success rate of the combination in the presence of metronidazole resistance (71%) and the low rate of tetracycline resistance. A different mix of resistant and sensitive H. pylori could have produced markedly different results. Worldwide tetracycline resistance appears to be low, so tetracycline may be an underutilized antibiotic for H. pylori infection. Our modification of the triple standard therapy (bismuth, tetracycline, and nitroimidazole) in which bismuth was replaced by a proton pump inhibitor was successful in 100% of patients with sensitive H. pylori. A previous study reported this combination was less effective but did not control for antimicrobial resistance and used a lower dose of tetracycline (500 mg tid) [20]. In conclusion, our results show that the efficacy rates of eradication therapy differ significantly in relation to pretreatment antimicrobial resistance or sensitive organisms. Although our study does not resolve the problem of the inclusion of antimicrobial evaluation before any eradication therapy, the problem of resistant H. pylori seems to be consistently increasing. This problem warrants new approaches, including cost-effectiveness analysis of standard therapies as opposed to resistance-determined antimicrobial drugs. A portion of this study was presented in abstract form at the Eleventh International Workshop on Gastroduodenal Pathology and Helicobacter pylori, Budapest, Hungary, September 2–5, 1998. This work was supported by the Institute of Internal Medicine, University of Sassari, Italy, and by a grant, Indagine Epidemiologica e Clinica sull’infezione da Helicobacter pylori in Sardegna e sulla patologia correlata, from the Regione Sardegna, Italy (Grant No. 459895). The authors thank Dr. Gioacchino Leandro for his contribution
to the statistical analysis, and Laura Spano and Suor Annunziata for their collaboration in patient assistance.
References 1 NIH Consensus Conference. Helicobacter pylori in peptic ulcer disease. JAMA 1994;272:65–9. 2 Penston JG, McColl KEL. Eradication of Helicobacter pylori: An objective assessment of current therapies. Br J Clin Pharmacol 1997;43:223–43. 3 Dixon JS. Helicobacter pylori eradication: Unraveling the facts. Scand J Gastroenterol 1995;30 (suppl 212):48–62. 4 Unge P. Review of Helicobacter pylori eradication regimens. Scand J Gastroenterol 1996;31 (suppl 215):74–81. 5 Megraud F. Resistance of Helicobacter pylori to antibiotics. Aliment Pharmacol Ther 1997;11 (suppl 1): 43–53. 6 Misiewic JJ, Tytgat GNJ, Goodwin CS. The Sidney System: A new classification of gastritis. J Hepatol Gastroenterol 1991;6:209–22. 7 National Committee for Clinical Laboratory Standards. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically: Approved standards [NCCLS publication M7-A3]. Villanova, PA: National Committee for Clinical Laboratory Standards, 1997. 8 Graham DY. A reliable cure for Helicobacter pylori infection? Gut 1995;37:154–6. 9 Graham DY, Dore MP. Variability in the outcome of treatment of Helicobacter pylori infection: A critical analysis. In: Hunt R, Tytgat GNJ, eds. Helicobacter pylori: Basic Mechanisms to Clinical Cure. Norwell: Kluwer Academic, 1998;426–40. 10 Fagan T. Quick basic program for exact and midP confidence intervals for a binomial proportion. Comput Biol Med 1996;26:263–7. 11 Blecker U, Gold BD. Treatment of Helicobacter pylori infection: A review. Pediatr Infect Dis J 1997;16:391–9. 12 Graham DY, Lew GM, Malaty HM, Evans DG, Evans Jr DJ, Klein PD, Alpert LC, Genta RM. Factors influencing the eradication of Helicobacter pylori with triple therapy. Gastroenterology 1992;102:493–6. 13 Graham DY, de Boer WA, Tytgat GN. Choosing the best anti–Helicobacter pylori therapy: Effect of antimicrobial resistance. Am J Gastroenterol 1996; 91:1072–6. 14 van de Hulst RW, Keller JJ, Rauws EAJ, Tytgat GN. Treatment of Helicobacter pylori infection: A review of the world literature. Helicobacter 1996;1:6–19. 15 Dore MP, Osato MS, Realdi G, Mura I, Graham DY, Sepulveda AR. Amoxicillin tolerance in Helicobacter pylori. J Antimicrob Chemother 1999; 43:47–54. 16 Dore MP, Piana A, Carta M, Atzei A, Are BM, Mura I, Massarelli G, Maida A, Sepulveda AR,
112 Graham DY, Realdi G. Amoxicillin resistance is one reason for failure of amoxicillin omeprazole treatment of Helicobacter pylori infection. Aliment Pharmacol Ther 1998;12:635–9. 17 Lind T, Veldhuyzen van Zanten S, Unge P, Spiller R, Bayerdorffer E, O’Morain C, Bardhan KD, Bradette M, Chiba N, Wrangstadh M, Cederberg C, Idstrom JP. Eradication of Helicobacter pylori using one-week triple therapies combining omeprazole with two antimicrobials: The MACH I study. Helicobacter 1996;1:138–44. 18 Nakata H, Itoh H, Nishioka S. Efficacy of lansopra-
Realdi et al. zole and amoxicillin in eradicating Helicobacter pylori: Evaluation using, 13C-UBT and monoclonal H. pylori antibody testing. J Clin Gastroenterol 1995;20 (suppl 2):S118–20. 19 Laine L, Estrada R, Trujillo M, Emami S. Randomized comparison of ranitidine bismuth citratebased triple therapies for Helicobacter pylori. Am J Gastroenterol 1997;92:2213–5. 20 McCarthy CJ, Collins R, Beattie S, Hamilton H, O’Morain C. Short report: Treatment of Helicobacter pylori–associated duodenal ulcer. Aliment Pharmacol Ther 1993;4:463–6.
