nature publishing group CLINICAL REVIEWS 65

REVIEW
Empiric Quadruple vs. Triple Therapy for Primary
Treatment of Helicobacter pylori Infection: Systematic
Review and Meta-Analysis of Efficacy and Tolerability
Jay Luther, MD1, Peter D.R. Higgins, MD, PhD1, Phillip S. Schoenfeld, MD1, Paul Moayyedi, MB ChB, PhD2, Nimish Vakil, MD3 and
William D. Chey, MD, AGAF, FACG, FACP1

OBJECTIVES: Recent treatment guidelines recommend two first-line therapies for Helicobacter pylori infection:
proton pump inhibitor (PPI), bismuth, tetracycline, and metronidazole (quadruple therapy) or PPI,
clarithromycin, and amoxicillin (triple therapy). We performed a systematic review and meta-analysis
to compare the efficacy and tolerability of these regimens as first-line treatment of H. pylori.

METHODS: A search of MEDLINE, EMBASE, Google Scholar, the Cochrane Central Register of Controlled
Trials, ACP Journal Club, the Database of Abstracts of Reviews of Effectiveness, Cochrane
Methodology Register, Health Technology Assessment Database, and abstracts from prominent
gastrointestinal scientific meetings was carried out. Randomized controlled trials (RCTs)
comparing bismuth quadruple therapy to clarithromycin triple therapy were selected for meta-
analysis. Two independent reviewers extracted data, using standardized data forms. Meta-analysis
was carried out with the metan command in Stata 10.1. Funnel plots and subgroup analyses
were carried out.

RESULTS: Nine RCTs (N = 1,679) were included. Although dosing regimens of clarithromycin triple
therapy were quite consistent between trials, dosing regimens varied considerably for bismuth
quadruple therapy. Bismuth quadruple therapy achieved eradication in 78.3% of patients,
whereas clarithromycin triple therapy achieved an eradication rate of 77.0% (risk ratio
(RR) = 1.002, 95% confidence interval (CI): 0.936 –1.073). There was moderate heterogeneity
and no evidence for significant publication bias. Subgroup analyses by study location, treatment
duration, and study population did not account for the heterogeneity. There were no statistically
significant differences in side effects yielded by quadruple vs. clarithromycin triple therapy
(RR = 1.04, 95% CI: 1.04 –1.14).

CONCLUSIONS: Quadruple and triple therapies yielded similar eradication rates as primary therapy for H. pylori
infection. Both therapies yielded suboptimal eradication rates. Patient compliance and side
effects are similar for quadruple and triple therapies.
Am J Gastroenterol 2010; 105:65–73; doi:10.1038/ajg.2009.508; published online 15 September 2009

INTRODUCTION serious clinical consequences of chronic H. pylori infection,

Helicobacter pylori infection affects more than half of the
world’s population and remains a significant cause of morbi-
dity and mortality. H. pylori has a critical role in the develop-
ment of chronic gastritis and peptic ulcer disease and has been
linked to the pathogenesis of gastric lymphoma and gastric
adenocarcinoma (1–5). Given the prevalence and potentially
the identification of highly effective and well-tolerated treat-
ment regimens is critically important.
Recent management guidelines from the Maastricht consen-
sus conference and the American College of Gastroenterology
recommend first-line treatment of H. pylori infection with
the combination of a proton pump inhibitor (PPI), bismuth,

1
Division of Gastroenterology, University of Michigan Medical Center, Ann Arbor, Michigan, USA; 2Gastroenterology Division, McMaster University, Hamilton, Ontario, Canada;
3
Division of Gastroenterology, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin, USA. Correspondence: William D. Chey, MD, AGAF, FACG,
FACP, GI Physiology Laboratory, University of Michigan Health System, 3912 Taubman Center, SPC 0362, Ann Arbor, Michigan 48109-0362, USA. E-mail: wchey@umich.edu
Received 26 February 2009; accepted 29 July 2009

© 2010 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 66 Luther et al.
metronidazole, and tetracycline (bismuth quadruple therapy)
for 10–14 days or with a PPI, clarithromycin, and amoxicil-
lin (clarithromycin triple therapy) for 7–14 days (6,7). Despite
REVIEW
these recent recommendations, concerns remain regarding
these treatment regimens. For example, there are growing
concerns about the efficacy of these regimens because of
possible increases in the prevalence of clarithromycin- and
metronidazole-resistant strains of H. pylori. The tolerability
of these regimens, in particular bismuth quadruple therapy,
also remains a concern among clinicians (8,9).
To further evaluate the efficacy and tolerability of
quadruple vs. clarithromycin triple therapy as first-line treat-
ment for H. pylori, we carried out a systematic review and
meta-analysis of randomized controlled trials comparing these
two treatment regimens.
METHODS
Search strategy
To find relevant articles for this review, we searched the fol-
lowing databases: MEDLINE, EMBASE, Google Scholar, the
Cochrane Central Register of Controlled Trials, ACP Journal
Club, The Database of Abstracts of Reviews of Effectiveness,
Cochrane Methodology Register, and Health Technology
Assessment Database. The following search terms were used:
“Helicobacter pylori,” “triple therapy,” “quadruple therapy,” “PPI,”
“omeprazole,” “lansoprazole,” “amoxicillin,” “clarithromycin,”
“bismuth,” “flagyl,” “metronidazole,” “tetracycline,” “eradication,”
and “treatment.” Boolean operators (AND, OR, NOT) were used
to narrow and widen the search results. The titles and abstracts
from the search results were examined closely and determined
to be suitable for potential inclusion into the study. In addition,
the references from selected articles were examined as a further
search tool, along with the abstracts, which were presented at
Digestive Disease Week and the American College of Gastroen-
terology Annual Scientific Meeting from 2003 to 2008.
Study selection
For inclusion in the meta-analysis, a study had to meet the
following: (i) randomized controlled trial; (ii) year published
after 1990; (iii) treatment with bismuth quadruple therapy
(metronidazole, bismuth-containing compound, tetracycline,
and PPI) and clarithromycin triple therapy (amoxicillin, clari-
thromycin, and PPI); (iv) same duration of treatment with bis-
muth quadruple and clarithromycin triple therapy; (v) method
of H. pylori diagnosis by urea breath test (UBT), rapid urease
test (RUT), histology, and/or fecal antigen testing; (vi) main
outcome measure as an intention-to-treat (ITT) eradication
rate; and (vii) eradication testing with UBT and/or histology
at least 4 weeks after completion of therapy.
Data extraction
To reduce reporting bias and error in data collection, two
independent reviewers (J.L. and W.D.C.) extracted data from
selected studies using standardized data extraction forms.
The American Journal of GASTROENTEROLOGY
These forms, created by the study team, included the following:
(i) authors, (ii) title, (iii) year of publication, (iv) journal, (v) study
design, (vi) inclusion and exclusion criteria, (vii) method by
which H. pylori infection was diagnosed and method of eradica-
tion testing, (viii) number (with sex differentiation if applicable)
and mean age of the study population, (ix) the number of patients
receiving bismuth quadruple therapy and within this group, the
number who succeeded and failed to eradicate H. pylori, (x) the
number of patients receiving clarithromycin triple therapy and,
within this group, the number who succeeded and failed to eradi-
cate H. pylori, (xi) the number of patients compliant and non-
compliant in the bismuth quadruple therapy group, (xii) number
of patients compliant and noncompliant in the clarithromycin
triple therapy group, (xiii) the number of patients reporting side
effects and absence of side effects in the bismuth quadruple ther-
apy group, and (xiv) number reporting side effects and absence of
side effects in the clarithromycin triple therapy group. If needed,
authors were contacted regarding specific questions relating to
their study. The independent reviewers collaborated once data
extraction was complete and discussed any discrepancies, at
which time a consensus was reached.
Statistical analysis
The primary study outcomes for the meta-analysis included
the following: (i) eradication rate of bismuth quadruple
therapy compared with clarithromycin triple therapy reported
as ITT, (ii) compliance rate of bismuth quadruple therapy vs.
clarithromycin triple therapy, and (iii) incidence of side effects
associated with bismuth quadruple therapy vs. clarithromy-
cin triple therapy. Risk ratio (RR) was used as a measure of
association between variables describing efficacy of bismuth
quadruple vs. clarithromycin triple therapy, and a 95% confi-
dence interval (95% CI) was calculated based on a random-
effects model as described by Mantel–Haenszel. Meta-analysis
was carried out with the metan command in Stata 10.1
(StataCorp LP, College Station, TX). Funnel plots and sub-
group analyses were carried out.
An I2 statistic was used to measure the proportion of incon-
sistency in individual studies that could not be explained by
chance (10). An I2 value >70% would suggest significant hetero-
geneity and preclude summation of individual study results.
Any heterogeneity identified would prompt subgroup analysis
in attempt to explain these findings. Furthermore, the cumula-
tive effect over time was analyzed using statistical software. This
same method was applied to assess compliance rates and the
incidence of adverse events between the two study groups.
RESULTS
Summaries of the included studies
Our initial search strategy yielded 212 potential articles for
inclusion. After detailed analysis of selected articles, 9 rando-
mized controlled trials (11–19) with 1,679 patients fulfilled
the inclusion criteria for the meta-analysis (Figure 1). The
most common reasons for exclusion from the meta-analysis
VOLUME 105 | JANUARY 2010 www.amjgastro.com

212 Studies identified from
search strategy
193 Studies were excluded (title
and abstract suggested article
was not appropriate)
19 Studies retrieved for detailed
evaluation
10 Studies excluded
Inappropriate drug regimen: 7
Different treatment duration: 2
Second-line treatment: 1
9 Studies included in
the meta-analysis
Figure 1. Flow diagram of studies identified in the systematic review and
meta-analysis.
included treatment regimens offered as second-line treat-
ment, different duration of treatment regimens, and regimens
composed of medications inconsistent with traditional triple
or bismuth quadruple therapy. Characteristics and results of
the included studies are summarized in Tables 1 and 2,
respectively.
Gomollon et al. (11) studied 48 patients receiving bismuth
quadruple therapy and 49 patients receiving clarithromycin
triple therapy in a prospective, randomized study from Spain
published in 2000. Patients with endoscopically proven peptic
ulcer disease in addition to H. pylori, diagnosed by both RUT
and histology, were included in the study. Exclusion criteria
included pregnancy, allergy to any study medications, chronic
nonsteroidal anti-inflammatory drug use, severe esophagitis,
previous H. pylori treatment, or recent gastrointestinal bleed-
ing. Eradication was confirmed with 13C-UBT or biopsy-based
testing 2 months after treatment. A total of 68.7% of patients
receiving bismuth quadruple therapy and 81.6% of patients
receiving clarithromycin triple therapy were cured. It can be
noted that, patients in the quadruple regimen arm received
metronidazole 250 mg t.i.d. (three times a day), a lower dose as
compared with other included studies.
Katelaris et al. (12) conducted a multicenter study in
New Zealand and Australia, which randomized 110 patients
to conventional bismuth quadruple therapy and 104 patients
to clarithromycin triple therapy. Data from a third treatment
arm consisting of the combination of bismuth, tetracycline,
and metronidazole were not included in this analysis. Inclu-
sion criteria were age over 18 years, diagnosis of H. pylori by
RUT, histology, or 13C-UBT, and endoscopy with no evidence of
peptic ulcer disease or esophagitis. Exclusion criteria included
any previous treatment for H. pylori or the use of PPIs, anti-
biotics, bismuth, or anti-inflammatories within 30 days of the
© 2010 by the American College of Gastroenterology
Quadruple vs. Triple Therapy for H. pylori 67
study. At 8 weeks after treatment, H. pylori status was deter-
mined by 13C-UBT. Pretreatment resistance testing in selected
patients showed significantly higher metronidazole resistance
REVIEW
compared with clarithromycin resistance in both treatment
groups. Bismuth quadruple therapy yielded an eradication rate
of 82%, whereas 78% receiving clarithromycin triple therapy
were successfully cured of H. pylori. It is notable that this study
reported significant protocol violations within each group, the
most common of which was antibiotic use between the 2- and
8-week follow-up visits.
Laine et al. (13) performed a multi-center study in the United
States and Canada, in which 138 patients were randomized to
receive bismuth quadruple therapy and 137 patients to receive
clarithromycin triple therapy. Inclusion criteria included docu-
mented duodenal ulcer detected by endoscopy and H. pylori
infection diagnosed by 13C-UBT and histology. Recent gastro-
intestinal bleeding, previous treatment for H. pylori, use of
antibiotics or bismuth within 30 days of enrollment, chronic
nonsteroidal anti-inflammatory drug use, or any contraindica-
tion to the study medications constituted the exclusion criteria.
H. pylori cure was documented with a 13C-UBT after 29 and 57
days of treatment completion. Both UBTs had to be negative for
treatment to be considered successful. The populations studied
showed significantly greater resistance to metronidazole than
clarithromycin. A total of 87.7% of patients receiving bismuth
quadruple therapy and 83.2% of patients receiving clarithro-
mycin triple therapy achieved successful H. pylori eradication.
In a study from Spain, Calvet et al. (14) randomized 168
patients to bismuth quadruple therapy and 171 patients to
clarithromycin triple therapy. Patients with a diagnosis of
peptic ulcer disease documented by endoscopy along with
H. pylori infection established by 13C-UBT, RUT, or histol-
ogy were included. Exclusion criteria included patients < 18
years of age, previous H. pylori treatment, antibiotic use within
4 weeks of enrollment, and/or previous ulcer surgery. Post-
treatment H. pylori status was established at least 2 months after
the completion of treatment (average 2.75 months) with either
biopsy-based testing or a 13C-UBT. Eradication was achieved in
83% of the bismuth quadruple therapy group and 77% of the
clarithromycin triple therapy group. As a significant proportion
of the study population presented with bleeding ulcers, many
patients had received anti-secretory treatment before enroll-
ment. In addition, this study was unusual, in that it allowed the
enrollment of patients with penicillin allergy. Six penicillin-
allergic patients randomized to clarithromycin triple therapy
were automatically counted as treatment failures.
Uygun et al. (15) conducted a single-center study in which 120
patients received bismuth quadruple therapy and 120 patients
received clarithromycin triple therapy. Study participants had
non-ulcer dyspepsia and H. pylori infection documented by
both UBT and histology. Post-treatment H. pylori status was
determined with a UBT 6 weeks after the completion of treat-
ment. The eradication rate was 70.0% with bismuth quadruple
therapy and 57.5% with clarithromycin triple therapy. This
study was carried out in Turkey where previous studies have
The American Journal of GASTROENTEROLOGY
 68 Luther et al.

Table 1. Study characteristics

Treatment
REVIEW

Author Year Location Triple-drug regimen Quadruple-drug regimen duration (days)

Gomollon et al. 2000 Spain Omeprazole 20 mg, b.i.d. Omeprazole 20 mg, b.i.d. 7
(11) Amoxicillin 1 g, b.i.d. Tetracycline 500 mg, t.i.d.
Clarithromycin 500 mg, b.i.d. Bismuth subcitrate 120 mg, t.i.d.
Metronidazole 250 mg, t.i.d.
Katelaris et al. 2002 Australia/ Pantoprazole 40 mg, b.i.d. Pantoprazole 40 mg, b.i.d. 7
(12) New Zealand Amoxicillin 1 g, b.i.d. Tetracycline 500 mg, q.i.d.
Clarithromycin 500 mg, b.i.d. Bismuth subcitrate 108 mg, q.i.d.
Metronidazole 200 mg t.i.d./400 mg q.h.s.
Calvet et al. 2002 Spain Omeprazole 20 mg, b.i.d. Omeprazole 20 mg, b.i.d. 7
(14) Amoxicillin 1 g, b.i.d. Tetracycline 500 mg, t.i.d.
Clarithromycin 500 mg, b.i.d. Bismuth subcitrate 120 mg, t.i.d.
Metronidazole 500 mg, t.i.d.
Mantzaris et al. 2002 Greece Omeprazole 20 mg, b.i.d. Omeprazole 20 mg, b.i.d. 10
(17) Amoxicillin 1 g, b.i.d. Tetracycline 500 mg, q.i.d.
Clarithromycin 500 mg, b.i.d. Bismuth subcitrate 120 mg, q.i.d.
Metronidazole 500 mg, t.i.d.
Pai et al. 2003 India Lansoprazole 30 mg, b.i.d. Lansoprazole 30 mg, b.i.d. 10
(16) Amoxicillin 500 mg, q.i.d. Tetracycline 500 mg, q.i.d.
Clarithromycin 500 mg, b.i.d. Tri-potassium dicitrato bismuth 120 mg, q.i.d.
Metronidazole 400 mg, t.i.d.
Laine et al. 2003 United States/ Omeprazole 20 mg, b.i.d. Omeprazole 20 mg, b.i.d. 10
(13) Canada Amoxicillin 1 g, b.i.d. Tetracycline 375 mg, q.i.d.
Clarithromycin 500 mg, b.i.d. Bismuth biskalcitrate 420 mg, q.i.d.
Metronidazole 375 mg, t.i.d.
Jang et al. 2005 Korea PPI, b.i.d. PPI, b.i.d. 7
(18) Amoxicillin 1 g, b.i.d. Tetracycline 500 mg, q.i.d.
Clarithromycin 500 mg, b.i.d. Bismuth subcitrate 300 mg, q.i.d.
Metronidazole 500 mg, t.i.d.
Uygun et al. 2007 Turkey Lansoprazole 30 mg, b.i.d. Lansoprazole 30 mg, b.i.d. 14
(15) Amoxicillin 1 g, b.i.d. Tetracycline 500 mg, q.i.d.
Clarithromycin 500 mg, b.i.d. Bismuth subsalicylate 300 mg, q.i.d.
Metronidazole 500 mg, t.i.d.
Ching et al. 2008 United Kingdom Lansoprazole 30 mg, b.i.d. Lansoprazole 30 mg, b.i.d. 7
(19) Amoxicillin 1 g, b.i.d. Tetracycline 500 mg, q.i.d.
Clarithromycin 500 mg, b.i.d. Bismuth subcitrate 240 mg, b.i.d.
Metronidazole 500 mg, t.i.d.
b.i.d., twice a day; PPI, proton pump inhibitor; q.h.s., nightly; q.i.d, four times a day; t.i.d., three times a day.

documented high clarithromycin resistance rates (20) and low
eradication rates with clarithromycin-based clarithromycin
triple therapy (21,22).
In a multicenter Indian study, Pai et al. (16) treated 33 and
35 patients with quadruple and triple therapies, respectively.
The participants had H. pylori infection documented by RUT
and histology. Those unable to tolerate oral medication and
patients who had taken PPI, antibiotics, or bismuth within 2
weeks of enrollment endoscopy were excluded from the study.
H. pylori status was reassessed through endoscopy and biopsy
30 days after completion of treatment. A total of 73 and 83%
of the quadruple and clarithromycin triple therapy groups,
respectively, were successfully eradicated of their H. pylori
infection.
Mantzaris et al. (17) conducted a prospective, investigator-
blinded, single-center study involving patients with confirmed
active duodenal ulceration by endoscopy and H. pylori infec-
tion by RUT and histology. The study included 71 patients in
the bismuth quadruple therapy group and 78 patients in the
clarithromycin triple therapy group. Repeat endoscopy was
performed 10–12 weeks after the completion of treatment
to assess H. pylori status with RUT, histology, and immuno-
chemistry. A total of 82 and 86% of patients in the quadruple
and triple groups, respectively, were cured of H. pylori infec-
tion. This study was carried out in Greece, a country noted to
have very high rates of metronidazole resistance.
In a study from Korea, Jang et al. (18) randomized 74 patients
to bismuth quadruple therapy and 75 patients to clarithromycin

The American Journal of GASTROENTEROLOGY VOLUME 105 | JANUARY 2010 www.amjgastro.com

 Quadruple vs. Triple Therapy for H. pylori 69

Table 2. Study results

Author Therapy No. of patients No. of cured patients ITT (%) Compliance Side effects

REVIEW
Gomollon et al. (11) Triple 49 40 81.6 98% 40%
Quadruple 48 33 68.8 100% 42%
Katelaris et al. (12) Triple 134 104 78.0 97% NR
Quadruple 134 110 82.0 94% NR
Calvet et al. (14) Triple 171 132 77.0 94% 59%
Quadruple 168 139 83.0 91% 59%
Mantzaris et al. (17) Triple 78 61 78.2 96% NR
Quadruple 71 46 64.8 93% NR
Pai et al. (16) Triple 35 29 82.9 100% 11%
Quadruple 33 24 72.7 94% 15%
Laine et al. (13) Triple 137 114 83.2 94% 59%
Quadruple 138 121 87.7 91% 59%
Jang et al. (18) Triple 75 59 78.7 NR 7%
Quadruple 74 53 71.6 NR 10%
Uygun et al. (15) Triple 120 69 57.5 96% 11%
Quadruple 120 84 70.0 91% 13%
Ching et al. (19) Triple 50 46 92.00 100% 90%
Quadruple 44 40 91.00 86% 95%
ITT, intention to treat; NR, not reported.

triple therapy. Patients enrolled had peptic ulcer disease and
H. pylori proven with RUT and histology. Post-treatment
testing was with a 13C-UBT performed 4–6 weeks after the
completion of therapy. Eradication rates were 71.6 and
78.7% for the quadruple and clarithromycin triple therapy
groups, respectively. Twelve patients receiving bismuth
quadruple therapy and six patients in the clarithromycin
triple therapy group either did not complete therapy or
were lost to follow-up.
Ching et al. (19) from North Wales, UK, randomized 101
patients with peptic ulcer or non-ulcer dyspepsia and H. pylori
infection documented by at least two of the following: RUT,
histology, or culture to bismuth quadruple or clarithromycin
triple therapy. More patients in the bismuth quadruple ther-
apy than in the clarithromycin triple therapy group admitted
to smoking (36 vs. 20%). Some have suggested that smoking
might reduce the efficacy of H. pylori treatments (23). The
exclusion criteria included age < 18 or >75 years, sympto-
matic gallstones, use of antibiotics or bismuth compounds
1 month before inclusion, PPI within 1 week of enrollment,
heavy alcohol consumption, serious co-morbid illness, use
of medications interfering with study drugs, and allergy to
study medications. Eradication of infection was established
with a UBT two months after the completion of therapy. These
investigators reported eradication rates of 91 and 92% in the
bismuth quadruple and clarithromycin triple therapy groups,
respectively.
Meta-analysis of ITT eradication rates
In sum, bismuth quadruple therapy achieved an eradication
rate of 78.3%, whereas the eradication rate with clarithromycin
triple therapy was 77.0% (RR = 1.002, 95% CI: 0.936–1.073).
Our analysis did have moderate heterogeneity (I2 = 43.9%)
(Figure 2).
In an attempt to explain heterogeneity, we conducted mul-
tiple subgroup analyses (Table 3). We first stratified studies
by location, more specifically studies conducted in the East-
ern vs. Western hemisphere, thereby acknowledging different
antimicrobial resistance rates and the effects this may have on
treatment outcomes (24–29). Our analysis showed no signif-
icant difference in eradication rates between quadruple and
clarithromycin triple therapy within each group. Heterogene-
ity within the Western hemisphere group (I2 = 55.2%) exceeded
our baseline results. In addition, we stratified studies by treat-
ment duration, because varied success rates have been reported
based on number of days treated (30–32). We divided studies
into one of three groups: (a) 7 days of treatment, (b) 10 days of
treatment, and (c) 14 days of treatment. Once again no signif-
icant difference in eradication rates was seen between quad-
ruple and clarithromycin triple therapy within each group, yet

© 2010 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 70 Luther et al.

%
Study RR (95% CI) Weight
REVIEW

Gomollon et al. 0.84 (0.67, 1.06) 6.5

Calvet et al. 1.07 (0.96, 1.19) 16.3

Katelaris et al. 1.06 (0.94, 1.19) 14.6

Mantzaris et al. 0.83 (0.67, 1.02) 7.7

Laine et al. 1.05 (0.96, 1.16) 17.4

Pai et al. 0.88 (0.68, 1.14) 5.5

Jang et al. 0.91 (0.76, 1.10) 9.0

Uygun et al. 1.22 (1.00, 1.48) 8.5

Ching et al. 0.99 (0.87, 1.12) 14.2

Overall (I 2 = 43.9%, P =0.075) 1.00 (0.94, 1.07) 100.0

Note: Weights are from random effects.

0.66 1 1.5

Figure 2. Forest plot of quadruple vs. clarithromycin triple therapy. CI, confidence interval; RR, risk ratio.

Table 3. Subgroup analyses

Eradication rate Eradication rate

Participants Studies with clarithromycin with bismuth
Analysis (references) (n) (n) triple therapy quadruple therapy RR (95% CI) I 2 statisitic

All included studies (11 –19) 1,679 9 78.3 77 1.002 (0.936 –1.073) 43.9
Location: Eastern hemisphere 1,194 6 76.4 78.6 1.011 (0.925 –1.105) 55.2
(11,13-15,17,19)
Location: Western hemisphere 485 3 78.7 77.6 0.978 (0.868 –1.102) 28.4
(12,16,18)
Duration: 7 days 947 5 79.5 80.1 1.004 (0.933 –1.081) 26.3
(11,12,14,18,19)
Duration: 10 days (13,16,17) 492 3 81.6 78.9 0.936 (0.783 –1.119) 65.2
Peptic ulcer disease 1,077 6 79.8 78.2 0.959 (0.872 –1.056) 51.6
(11,13,14,16 –18)
Non-ulcer dyspepsia (12,15) 508 2 68.1 76.4 1.113 (0.969 –1.278) 37.4
CI confidence interval; RR, risk ratio.

heterogeneity within the 10-day treatment group (I2 = 65.2%)
exceeded our baseline analysis. Lastly, we stratified studies
based on population, more specifically studies investigating
ulcer vs. non-ulcer dyspeptic patients, because varying degrees
of treatment efficacy has been suggested between the two groups
(33). Our analysis did not show a significant difference between
quadruple and clarithromycin triple therapy eradication rates,
yet within the ulcer group we found significant heterogeneity
(I2 = 51.6%).
Eradication rates related to antimicrobial susceptibility were
reported by two studies (12,13) (Figure 3). Overall, bismuth
quadruple therapy achieved eradication in 89.4% of metronida-
zole-sensitive strains of H. pylori and 80.6% of metronidazole-
resistant strains. In comparison, clarithromycin triple therapy
eradicated 90.2% of clarithromycin-sensitive strains of H. pylori
and 22.2% of clarithromycin-resistant strains.
Cumulative analysis to assess changes in eradication rate over
time showed no effect on outcome measure, yet it did show

The American Journal of GASTROENTEROLOGY VOLUME 105 | JANUARY 2010 www.amjgastro.com

 Quadruple vs. Triple Therapy for H. pylori 71

100
Metro- Clarithro- side effects within each group. The overall incidence of side
sensitive sensitive
90 effects in patients receiving bismuth quadruple therapy was
Eradication rates (percentage)

Metro-
resistant
80 35.5%, whereas the incidence in patients receiving clarithro-

REVIEW
70 mycin triple therapy was 35.4%. No statistically significant
60 difference exists between each group (RR: 1.037 with 95% CI:
50
1.037–1.135).
40

30 Clarithro-
resistant DISCUSSION
20
On the basis of our analysis, quadruple and triple therapies
10
yielded similar eradication rates when used as first-line therapy
0
Quadruple therapy Triple therapy
for H. pylori infection. The results of our updated meta-analy-
sis are consistent with an older meta-analysis that compared
Figure 3. Intention-to-treat eradication rates achieved by quadruple and these treatment regimens (34). The similar eradication rates
triple therapies based on metronidazole and clarithromycin resistance, were maintained in subgroup analyses by location, duration,
respectively (data derived from Katelaris et al. (12) and Laine et al. (13)).
and treatment population. This analysis also confirms that
there is no difference in compliance rate or the incidence
of treatment-associated side effects between quadruple and
Funnel plot with pseudo-95% confidence limits
0 clarithromycin triple therapy.
Despite the similar efficacy and tolerability of the two regi-
mens, it is notable that both regimens yielded suboptimal
ITT eradication rates of < 80%. A key determinant of treat-
s.e. (logRR)

0.05
0.1
0.15
0.8 0.9 1.0 1.1
Relative risk (log scale)
Figure 4. Funnel plot analysis. Funnel plot, Begg’s test, and Egger’s test
showed no evidence of publication bias.
tightening 95% CIs over time. In studies conducted in 2005 and
thereafter, bismuth quadruple therapy achieved an eradication
rate of 74.4% compared with 79.9% in studies conducted before
2005. Furthermore, clarithromycin triple therapy achieved an
eradication rate of 71.0% in studies conducted in and after
2005, whereas in studies conducted before 2005, an eradication
rate of 79.5% was achieved.
There was no evidence to suggest significant publication bias
according to Begg’s test, Egger’s test, or funnel plot analysis
(Figure 4).
Meta-analysis of compliance and side effects
Seven of nine studies (11–14,16,17,19) included data on com-
pliance rates. Each study defined acceptable compliance differ-
ently varying from 75 to 100% of completed therapy. Overall,
the compliance rate with bismuth quadruple therapy was
92.6%, whereas the compliance rate for clarithromycin triple
therapy was 96.9%. No statistically significant difference exists
between the two groups (RR: 0.99 with 95% CI: 0.938–1.045).
All but two studies (12,17) included describe the incidence of
ment failure is the presence of antimicrobial resistance. Anti-
microbial resistance varies by geographical region and is highly
influenced by patterns of antimicrobial use within a popula-
tion. Only two of the studies included in our meta-analysis
reported data from pretreatment antimicrobial sensitivity test-
ing (12,13). Laine et al. (13) reported significantly higher rates
1.2 1.3 of metronidazole resistance than clarithromycin resistance
(39.6 vs. 11.3%). However, the clinical impact of metronida-
zole resistance on bismuth quadruple therapy appears to differ
from that of clarithromycin resistance on traditional clarithro-
mycin triple therapy. They also reported a modest decrease in
eradication rates from 87.7% among all patients treated with
bismuth quadruple therapy to 80.4% in patients with evidence
of pretreatment metronidazole-resistant H. pylori infection.
In contrast, the efficacy of traditional clarithromycin triple
therapy was profoundly affected by the presence of pretreat-
ment clarithromycin resistance (overall eradication = 83.2%,
eradication in patients with pretreatment clarithromycin
resistance = 21.4%). Similar results were reported by Katelaris
et al. (12) and Fishbach and Evans (35).
In an additional secondary analysis, treatment duration did
not influence the efficacy of quadruple or clarithromycin triple
therapy. It should be noted that of the nine studies included in
our meta-analysis, five used 7-day treatment regimens, three
used 10-day regimens, and only one used 14-day regimens.
This may explain why our results differed from another meta-
analysis that found that 14-day treatment regimens were more
effective than 7-day regimens (36).
When considering the results of this meta-analysis, several
potential limitations should be considered. Moderate hetero-
geneity was present and could not be accounted for by multiple
subgroup analyses. In addition, individual studies included in
our analysis differed in multiple respects. For example, studies

© 2010 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 72 Luther et al.
defined varying inclusion and exclusion criteria, used different
PPIs, and set forth different definitions of compliance. Relevant
to bismuth quadruple therapy, various different antimicrobial
REVIEW
dosing regimens and bismuth formulations were used, which
could have affected eradication rates, as increasing dose and
duration can improve efficacy (37). Unfortunately, the hetero-
geneity in the regimens included in this review does not allow
for the recommendation of an “optimal” bismuth quadruple
therapy regimen.
In conclusion, first-line quadruple and triple therapies yielded
similar eradication rates in the treatment of H. pylori infec-
tion. In addition, compliance rates and the incidence of side
effects were similar between the treatment regimens. Unfortu-
nately, both regimens yielded suboptimal ITT eradication rates
of < 80%. It would be ideal to base the choice of treatment regi-
men upon H. pylori susceptibility testing. The current means of
resistance testing, which requires endoscopy and gastric mucosal
biopsy, is both invasive and costly, and newer noninvasive meth-
ods, such as fecal testing (38), are not yet available in routine
clinical practice. On account of this, it is critically important
to consider and address issues that might enhance compliance
or reduce the likelihood of pretreatment antimicrobial resist-
ance. Stressing the importance of compliance and fully disclos-
ing potential side effects when prescribing eradication therapy
should be pursued in every patient. If a patient has previously
been treated with a macrolide antibiotic for any reason, par-
ticularly if treatment has been taken repeatedly, the likelihood
of pretreatment clarithromycin resistance is increased. From a
practical perspective, it is reasonable to recommend bismuth
quadruple or clarithromycin triple therapy in patients who are
not penicillin allergic and have not previously been treated with
a macrolide antibiotic. In a patient with known penicillin allergy
or previous treatment with a macrolide antibiotic (particularly
if the patient has received multiple courses of a macrolide), bis-
muth quadruple therapy would be a logical choice. Given these
results, efforts to develop and validate more effective alternative
treatment strategies are strongly encouraged.
ACKNOWLEDGMENTS
We thank Stephen George for his assistance with the initial
data searches and feedback during preparation of this paper.
CONFLICT OF INTEREST
Guarantor of the article: Jay Luther, MD.
Specific author contributions: Substantial contributions to
the intellectual content of the paper (conception and design,
data acquisition, data analysis and interpretation, drafting, and
critical revision of the paper), assistance in writing the first
draft, and significant contributions to all subsequent versions
of the paper: Jay Luther; substantial contributions to the
intellectual content of the paper (data analysis and interpreta-
tion, drafting, and critical revision of the paper): Peter D.
Higgins; substantial contributions to the intellectual content
of the paper (data analysis and interpretation, drafting, and
critical revision of the paper): Phillip S. Schoenfeld; substantial
The American Journal of GASTROENTEROLOGY
contributions to the intellectual content of the paper
(conception and design, data acquisition, data analysis and
interpretation, drafting, and critical revision of the paper):
Paul Moayyedi; substantial contributions to the intellectual
content of the paper (conception and design, data acquisition,
data analysis and interpretation, drafting, and critical revision
of the paper): Nimish Vakil; substantial contributions to the
intellectual content of the paper (conception and design, data
acquisition, data analysis and interpretation, drafting, and
critical revision of the paper), created and edited the paper,
and offered final approval before submission: William D. Chey.
The authors attest that they carried out all of the data analyses
and wrote the paper without professional assistance.
Financial support: Axcan provided an unrestricted grant
to the University of Michigan to assist with the literature
searches that led to the conception of this meta-analysis.
Potential competing interests: Jay Luther has no conflicts of
interest relevant to this study. Peter D.R. Higgins has no con-
flicts of interest relevant to this study. Phillip S. Schoenfeld
has no conflicts of interest relevant to this study.
Paul Moayyedi has acted as an independent medical con-
sultant for AstraZeneca and Janssen Ortho. He is on the
speakers’ bureau for AstraZeneca, Janssen Ortho, Nycomed,
and Esai. He holds a Chair that is funded partly through an
unrestricted donation from AstraZeneca Canada to McMas-
ter University. Nimish Vakil is a consultant for AstraZeneca,
Axcan, Orexo, Xenoport, Takeda, and Merlion and has
received research grants from AstraZeneca, Addex, Dynogen,
Forest Labs, and Xenoport. He holds stock in Meridian
and Orexo and has provided expert testimony on behalf
of AstraZeneca. William D. Chey is a consultant for AGI,
Axcan, AstraZeneca, Ironwood, McNeil, Proctor &
Gamble, Prometheus, Salix, Smart Pill Corporation, Takeda,
and Xenoport and has been a speakers’ bureau member
for Axcan, Prometheus, Salix, and Takeda.
REFERENCES
1. Ford AC, Delaney BC, Forman D et al. Eradication therapy in Helicobacter
pylori positive peptic ulcer disease: systematic review and economic analy-
sis. Am J Gastroenterol 2004;99:1833–55.
2. Chen LT, Lin JT, Tai JJ et al. Long-term results of anti-Helicobacter pylori
therapy in early-stage gastric high grade transformed MALT lymphoma.
J Natl Cancer Inst 2005;97:1345–53.
3. Morgner A, Bayerdorffer E, Neubauer A et al. Malignant tumors of the
stomach. Gastric mucosa-associated lymphoid tissue lymphoma and
Helicobacter pylori. Gastroenterol Clin North Am 2000;29:593–607.
4. Malfertheiner P, Sipponen P, Naumann M et al. Helicobacter pylori eradica-
tion has the potential to prevent gastric cancer: a state-of-the-art critique.
Am J Gastroenterol 2005;100:2100–15.
5. Saad RJ, Chey WD. Treatment of Helicobacter infection in 2006. Gastroen-
terol Hepatol Ann Rev 2006;1:30–5.
6. Chey WD, Wong BCY. American College of Gastroenterology guideline
on the management of Helicobacter pylori infection. Am J Gastroenterol
2007;102:1808–25.
7. Malfertheiner P, Megraud F, O’Morain C et al. Current concepts in the
management of Helicobacter pylori infection: the Maastricht III Consensus
Report. Gut 2007;56:772–81.
8. Graham DY, Lew GM, Malaty HM et al. Factors influencing the eradication
of Helicobacter pylori with clarithromycin triple therapy. Gastroenterology
1992;102:493–6.
VOLUME 105 | JANUARY 2010 www.amjgastro.com

9. Isakov V, Domareva I, Koudryavtseva L et al. Furazolidone-based triple
“rescue therapy” vs. quadruple “rescue therapy” for the eradication of
Helicobacter pylori resistant to metronidazole. Aliment Pharmacol Ther
2002;16:1277–82.
10. Higgins JP, Thompson SG, Deeks JJ. Measuring inconsistency in
meta-analysis. BMJ 2003;327:557–60.
11. Gomollon F, Valdeperez J, Garuz R et al. Análisis de coste-efectividad de
dos estrategias de erradicación de Helicobacter pylori: resultados de un
estudio prospectivo y aleatorizado en atención primaria. Med Clin (Barc)
2000;115:1–6.
12. Katelaris PH, Forbes GM, Talley NJ et al. A randomized comparison of
quadruple and triple therapies for Helicobacter pylori eradication. Gastro-
eneterology 2002;123:1763–9.
13. Laine L, Hunt RH, El-Zimaity H et al. Bismuth-based bismuth quadruple
therapy using a single capsule of bismuth biskalcitrate, metronidazole, and
tetracycline given with omeprazole versus omeprazole, amoxicillin, and
clarithromycin for eradication of Helicobacter pylori in duodenal ulcer
patients: a prospective, randomized, North American trial. Am J
Gastroenterol 2003;98:562–7.
14. Calvet X, Ducons J, Guardiola J et al. One-week triple vs. bismuth
quadruple therapy for Helicobacter pylori infection—a randomized trial.
Aliment Pharmacol Ther 2002;16:1261–7.
15. Uygun A, Kadayifci A, Safall M et al. The efficacy of bismuth containing
bismuth quadruple therapy as a first line treatment option for Helicobacter
pylori. J Digest Dis 2007;8:211–5.
16. Pai CG, Thomas CP, Biswas A et al. Bismuth quadruple therapy for initial
eradication of Helicobacter pylori in peptic ulcer: comparison with
clarithromycin triple therapy. Indian J Gastroenterol 2003;22:85–7.
17. Mantzaris GJ, Petraki K, Archavlis E et al. Omeprazole clarithromycin
triple therapy versus omeprazole bismuth quadruple therapy for healing
duodenal ulcer and eradication of Helicobacter pylori infection: a 24 month
follow-up study. Eur J Gastroenterol Hepatol 2002;14:1237–43.
18. Jang HJ, Choi MH, Kim YS et al. Effectiveness of triple and bismuth quad-
ruple therapy for Helicobacter pylori eradication. Korean J Gastroenterol
2005;46:368–72.
19. Ching SS, Sabanathan S, Jenkinson LR. Treatment of Helicobacter pylori
in surgical practice: a randomised trial of triple versus bismuth quad-
ruple therapy in a rural district general hospital. World J Gastroenterol
2008;14:3855–60.
20. Simsek H, Balaban YH, Gunes DD et al. Alarming clarithromycin
resistance of Helicobacter pylori in Turkish population. Helicobacter
2005;10:360–1.
21. Uygun A, Ates Y, Erdil A et al. Efficacy of omeprazole plus two antimicrobi-
als for the eradication of Helicobacter pylori in a Turkish population.
Clin Ther 1999;21:1539–48.
22. Uygun A, Kadayifci A, Yesilova Z et al. Recent success of pantoprazole- or
lansoprazole-based clarithromycin plus amoxicillin treatment in the
© 2010 by the American College of Gastroenterology
Quadruple vs. Triple Therapy for H. pylori 73
eradication of Helicobacter pylori. Turk J Gastroenterol 2004;
15:219–24.
23. Graham DY, Shiotani A. New concepts of resistance in the treatment
of Helicobacter pylori infection. Nat Clin Pract Gastroenterol Hepatol
REVIEW
2008;5:321–31.
24. Gomollon F, Ducons JA, Gimeno L et al. The ideal therapy must be defined
in each geographical area: experience with a bismuth quadruple therapy in
Spain. Helicobacter 1998;3:110–4.
25. Van der Hulst RWM, van der Ende A, Horman A. Influence of metronida-
zole resistance on efficacy of bismuth quadruple therapy for Helicobacter
pylori eradication. Gut 1998;42:166–9.
26. Houben M, van de Beek D, Hensen E et al. A systematic review of
Helicobacter eradication therapy—the impact of antimicrobial resistance
on eradication rates. Aliment Pharmacol Ther 1999;13:1047–55.
27. Mégraud F. H. pylori antibiotic resistance: prevalence, importance, and
advances in testing. Gut 2004;53:1374–84.
28. Glupczynski Y, Megraud F, Lopez-Brea M et al. European multicentre
survey of in vitro antimicrobial resistance in Helicobacter pylori. Eur J Clin
Microbiol Infect Dis 2001;20:820–3.
29. Kim JM, Kim JS, Kim N et al. Distribution of antibiotic MICs for
Helicobacter strains over a 16-year period in patients from Seoul, Korea.
Antimicrob Agents Chemother 2004;48:4843–8.
30. Vakil N, Lanza F, Schwartz H et al. Seven-day therapy for Helicobacter
pylori in the United States. Aliment Pharmacol Ther 2004;20:99–107.
31. Laine L, Estrada R, Trujillo M et al. Randomized comparison of differing
periods of twice a day clarithromycin triple therapy for the eradication of
Helicobacter pylori. Aliment Pharmacol Ther 1996;10:1029–33.
32. Zagari RM, Bianchi-Porro G, Fiocca R et al. Comparison of 1 and 2 weeks
of omeprazole, amoxicillin and clarithromycin treatment for Helicobacter
pylori eradication: the HYPER Study. Gut 2007;56:475–9.
33. Broutet N, Tchamgoue S, Pereira E et al. Risk factors for failure of
Helicobacter pylori therapy—results of an individual data analysis of
2751 patients. Aliment Pharmacol Ther 2003;17:99–109.
34. Gene E, Calvet X, Azagra R. Triple versus bismuth quadruple therapy for
treating Helicobacter pylori infection: a meta-analysis. Aliment Pharmacol
Ther 2003;17:1137–43.
35. Fishbach L, Evans EL. Meta-analysis: the effect of antibiotic resistance
status on the efficacy of triple and quadruple first-line therapies for
Helicobacter pylori. Aliment Pharmacol Ther 2007;26:343–57.
36. Calvet X, Garcia N, Lopez T et al. A meta-analysis of short versus long
therapy with a proton pump inhibitor, clarithromycin and either
metronidazole or amoxycillin for treating Helicobacter pylori infection.
Aliment Pharmacol Ther 2000;14:603–9.
37. Megraud F, Lamouliatte H. Review article: the treatment of refractory
H. pylori infection. Aliment Pharmacol Ther 2003;17:1333–43.
38. Fontana C, Favaro M, Pietroiusti A et al. Detection of clarithromycin-
resistant Helicobacter pylori in stool samples. J Clin Microbiol 2003;41:3636–40.
The American Journal of GASTROENTEROLOGY