COMMENTARIES
The Toronto
Helicobacter
pylori Consensus
in Context
G uideline development uses a
formal series of steps that are
time, resource, and labor intensive to
produce recommendations that are
likely to be trustworthy.1 The authors
of the Toronto Consensus for the
Treatment of Helicobacter pylori,
published in this issue of Gastroenter-
ology,2 followed such a rigorous
process for producing their recom-
mendations. Lack of important stake-
holders (eg, primary care, infectious
disease) on the panel was a
shortcoming. In addition, systematic
reviews were restricted to studies
published from 2008 onward.
Although the restriction was used to
avoid relying on older eradication data
when resistance rates may have been
lower, studies providing useful infor-
mation on achievable eradication rates
with different H pylori regimens may
have been missed and a formal
assessment of change in outcomes over
time could not be done. Nevertheless,
the process produced recommenda-
tions that should impact practice.
The consensus panel made strong
recommendations that first-line ther-
apy should be a 14-day course of
quadruple therapy—either traditional
bismuth-containing therapy or non–
bismuth-containing “concomitant”
therapy. The panel also strongly rec-
ommended that proton pump inhibitor
triple therapy, currently the most
widely used regimen, should not be
used unless physicians know that the
clarithromycin resistance is <15% or
the eradication rate is >85% in
their local population. They cite the
increase in clarithromycin resistance
and the dramatically lower eradication
rates with triple therapy in patients
with clarithromycin-resistant vs
clarithromycin-susceptible strains as
reasons to restrict triple therapy. The
consensus panel also strongly recom-
mended that the choice of first-line
therapy consider local antibiotic resis-
tance patterns and eradication rates,
citing a systematic review revealing
eradication rates with susceptibility-
based therapy of 90%–95% by
intention-to-treat analysis and 97%–
98% by per-protocol analysis in 4 of
the 5 component studies—significantly
higher than eradication rates with tri-
ple therapy and no susceptibility
testing.3
The last 2 recommendations lead to
a conundrum for clinicians. Unlike
most infectious diseases, culture, and
susceptibility testing for H pylori is
rarely performed in standard practice
and reliable recent data about suscep-
tibility are generally not available. This
may be due to several factors,
including that gastroenterologists
rather than infectious disease physi-
cians “own” H pylori and concerns
about the time and the difficulty
required for H pylori culture versus
other standard bacterial cultures.
Given the very limited data on current
resistance rates for many parts of the
world, including individual pop-
ulations in the United States, physi-
cians often have little local information
to guide their choice of therapy other
than their anecdotal experience with
current therapies.
Toronto H pylori
Consensus
Recommendations
in Context
Considered as an infectious disease,
H pylori is conceptually easy to treat.
Treatment success requires (1) use of
antimicrobials to which the organism
is susceptible; (2) use of antimicrobials
that reach all the niches occupied by
the organism; (3) depending on the
antimicrobial, adjuvants to increase
the pH so that the antimicrobial be-
comes more effective and so that
resting organisms divide; and (4)
established details of the regimen,
including drugs, doses, formulations,
frequency of administration, adminis-
tration in relation to meals, and dura-
tion of therapy. The most common
causes of treatment failure are antibi-
otic resistance and lack of patient
adherence to the regimen. Because not
only do H pylori strains vary, but so too
do the humans receiving treatment (eg,
CYP2C19 polymorphisms, allergies,
medication-induced symptoms, antibi-
otic history), the optimal regimen for
an individual or population may also
vary in relation to drugs and doses of
drugs used.
H pylori may be the only common
disease that gastroenterologists treat
with medications for which there is no
placebo response. For infectious dis-
eases, studies designed to confirm that
a new therapy achieves a prespecified
threshold of success (eg, 95%) do not
require a comparator.4 Comparative
studies are done as noninferiority tri-
als to ensure that new regimens are
not worse than the best current
regimen.4
Infectious disease therapy is
fundamentally susceptibility based.
Population eradication rates may vary
widely depending on resistance rates
in the study populations. The lack of
susceptibility data in published H
pylori “trial and error” treatment
studies provides population mean
results that reflect the sum of treat-
ment success with susceptible in-
fections plus the treatment success
with resistant infections.4 Figure 1
illustrates how the eradication rates
of a specific regimen such as triple
therapy vary widely across different
populations based on underlying
antibiotic susceptibility, despite no
change in efficacy in eradicating sus-
ceptible and resistant H pylori strains.
Given an expected 95% eradication
rate for individuals with susceptible
strains and 15% with resistant
strains, the population eradication
rate changes markedly across pop-
ulations based solely on the propor-
tion with resistant strains: from 95%
with no resistance to 75% with a 25%
resistance rate to 55% with a 50%
resistance rate (Figure 1).
Thus, comparison of a treatment
regimen across studies is problematic
and eradication rates likely represent
differences in the resistant rates of the
population studied rather than differ-
ences in the underlying efficacy of a
regimen. Aggregating studies as
required in meta-analyses may not be
Gastroenterology 2016;151:9–12
COMMENTARIES
Figure 1.Helicobacter pylori eradication rates with proton pump inhibitor triple
therapy in populations with varying proportions of clarithromycin resistance. Based
on a theoretical population where the cure rate is 95% with susceptible infections
and 15% with resistant infections.
useful in determining the efficacy of
the regimen itself. Rather, pooled es-
timates of eradication rates and the
precision around the estimates vary
greatly based on the variation in
resistance rates in the component
studies of the meta-analysis. If marked
variation in resistance rates exists
across component studies, heteroge-
neity in study results potentially may
be so great that aggregating the data to
provide a pooled estimate is of ques-
tionable utility. For example, using
data in Figure 1, if one-half of the
component studies in a meta-analysis
of triple therapy had near zero clari-
thromycin resistance (eradication rate
of 95%) and one-half the component
studies had around 50% resistance
(eradication rate of 55%), the pooled
eradication rate of 75% would provide
limited guidance in choosing an H
pylori therapy.
There are now sufficient data
regarding treatment outcomes in rela-
tion to susceptibility that one can
predict the eradication rate of a ther-
apy based on susceptibility data.4,5 The
details are well-described in the recent
literature, which also includes the for-
mulas, a web site, and a decision model
and sensitivity analysis based on the
effectiveness in relation to antibiotic
susceptibility.5–7 A number of different
regimens when prescribed to patients
with susceptible infections will reliably
achieve near 100% cure rates among
those who adhere to the regimen.8
The susceptibility-based approach
helps to explain why treatments
10
succeed and fail and has been reported
to be cost saving.9 Given high rates of
antibiotic-resistant strains in children,
susceptibility-based therapy has also
been suggested in the pediatric
population.10,11
The 14-day regimens of concomi-
tant and bismuth quadruple therapy
recommended by the Toronto
Consensus are appropriate as empiric
therapies in the absence of information
on susceptibility testing. It is important
to note that use of 4-drug non-bismuth
quadruple therapies promotes over-
prescribing of antibiotics because
those with susceptible strains receive
an antibiotic that is not required to
achieve eradication. For example, if
only 20% of patients who receive
concomitant therapy have clari-
thromycin resistance, then 80% will
receive metronidazole unnecessarily.
Figure 2.Example of one possible set of recommendations for susceptibility-based
first-line Helicobacter pylori therapy. This example is based on the premises that it
is preferable to minimize the use of unnecessary antibiotics and that proton pump
inhibitors, amoxicillin, and third drug triple therapies are usually better tolerated
than bismuth quadruple therapy.
Figure 2 is an example of how using
susceptibility to guide therapy could
reduce or eliminate the need to pre-
scribe unnecessary antimicrobials.
The reason for success with
concomitant therapy is that it is under-
mined only by dual metronidazole-
clarithromycin resistance, which occurs
in <5% of treatment-naïve patients in
the United States.8,12 Bismuth
quadruple therapy also performs well
as an empiric therapy because in vitro
metronidazole resistance does not
appear to markedly reduce efficacy, at
least when given for 14 days with
1500 mg of metronidazole.5,13–16 In
addition, resistance to tetracycline,
the other antibiotic in bismuth
quadruple therapy, is <1% in most of
the world.14,15 However, the details of
the optimum bismuth regimen remain
unclear. For example, is bismuth twice
a day equivalent to 4 times daily?
Is tetracycline twice a day equivalent
to 4 times daily? Is 800 mg of metro-
nidazole twice daily equal to 400 mg 4
times daily? Is metronidazole resistance
actually overcome or is the benefit
owing to bismuth killing the remaining
metronidazole-resistant organisms?
Finally, tetracycline is unavailable
currently in many countries, and
agents such as doxycycline cannot sub-
stitute effectively. Tetracycline is avail-
able in a combination product of
bismuth, tetracycline, and metronida-
zole (PYLERA, Aptalis Pharma US, Bir-
mingham, AL), although this product is
approved and packaged for only 10-day
therapy and is expensive (approxi-
mately $600 for 14 days).

The consensus group also recom-
mended a 14-day fluoroquinolone tri-
ple therapy for salvage therapy.
Unfortunately, fluoroquinolone resis-
tance has continued to increase with
the widespread use of fluo-
roquinolones for a variety of infections
(eg, resistance of 31% in Houston,
14.5% in Europe12,17–19) and, as noted
by the Toronto Consensus panel, pa-
tients with resistant strains have
markedly lower eradication rates.
Therefore, this recommendation is
likely already out of date for treating
physicians who lack information on
fluoroquinolone resistance in the pa-
tient or local population.5,8
The current problem with guiding
therapy using H pylori susceptibility
testing is that this testing is not readily
available, although local laboratories
and hospitals provide culture and
susceptibility testing and/or
molecular-based susceptibility testing
for every other common pathogen.
Ideally, such information would be
available readily to guide individual
decisions for H pylori therapy. How-
ever, in the absence of susceptibility
testing for individual patients, knowl-
edge of local resistance patterns al-
lows rational and regularly updated
empiric therapy. This is especially
important as most diagnostic testing
for H pylori is likely not biopsy based,
but rather is performed by non-
gastroenterologists using non-
endoscopic methods (eg, evaluation of
uninvestigated dyspepsia).
Moving Forward in the
Treatment of H pylori
Regardless of past impediments,
we believe it is important to increase
susceptibility testing and develop
methods to publicly communicate
current antibiotic resistance patterns
in local populations to better inform
the choice of appropriate H pylori
regimens for our patients. Such ef-
forts should allow us to improve on
the unacceptably low eradication
rates commonly achieved with many
therapies and potentially simplify the
complex therapies recommended at
present. A Helicobacter pylori Anti-
microbial Resistance Monitoring
Project was previously established
under the auspices of the Centers for
Disease Control and Prevention and
last reported regional trends of anti-
microbial resistance in H pylori iso-
lates from 11 hospitals in the United
States from 1998 to 2002.20 Suscep-
tibility testing will need to be more
widely available at local laboratories
and hospitals to accomplish this goal.
Mechanisms to communicate local
resistance patterns will also need to
be established. Support of our pro-
fessional societies is needed to
establish widely available suscepti-
bility testing and mechanisms to
communicate local patterns of anti-
biotic resistance.
Hopefully, simplified and improved
H pylori therapy can be developed
based on the concepts for designing
successful treatment discussed above.
For example, it is hypothesized that if
the pH of the niches occupied by H
pylori can be increased to 6, H pylori
will leave the resting or dormant state
and become highly susceptible to an-
timicrobials such as amoxicillin, for
which resistance is uncommon and not
increasing.4,21 The use of vonoprazan
and other potassium competitive acid
blockers, or of frequent high-dose
proton pump inhibitors, offers the
possibility that a simple dual therapy
may be able to reliably eradicate most
H pylori infections (discussed in detail
in16).
The recommendations of the Tor-
onto Consensus panel regarding use of
14-day, 4-drug empiric regimens and
avoidance of proton pump inhibitor
triple therapies when information on
resistance is not available are reason-
able and consistent with recommen-
dations resulting from analyses based
on susceptibility data.4–8 Broad guide-
line recommendations and pooled re-
sults from meta-analyses of
comparative studies are not sufficient
to describe all of the nuances a clini-
cian must consider in judging whether
a particular regimen is suitable for a
specific patient: for instance, local
resistance patterns if known, region of
origin/residence, regional antimicro-
bial use, patient’s antibiotic history and
allergies, condition necessitating H py-
lori therapy, cost, and availability of
medications.4–8 Interpreting studies of
an intervention (H pylori therapy) is
COMMENTARIES
also problematic when a key factor
(resistance) impacting the outcome of
interest (eradication) is unknown. We,
therefore, agree with the Toronto
Consensus panel that better informa-
tion on resistance patterns and eradi-
cation rates in local populations is
critical for choosing regimens that are
effective and also potentially less
complex to reduce overuse of antibi-
otics and improve tolerability and pa-
tient adherence.
DAVID Y. GRAHAM
Michael E. DeBakey VA Medical Center
and Baylor College of Medicine
Houston, Texas
LOREN LAINE
Yale School of Medicine
New Haven and
VA Connecticut Healthcare
West Haven, Connecticut
References
1. Institute of Medicine. Clinical prac-
tice guidelines we can trust.
Washington, DC: The National
Academies Press, 2011.
2. Fallone CA, Chiba N, van Zanten SV,
et al. The Toronto consensus for the
treatment of Helicobacter pylori
infection in adults. Gastroenter-
ology 2016;151:51–69.
3. Wenzhen Y, Yumin L, Quanlin G,
et al. Is antimicrobial susceptibility
testing necessary before first-line
treatment for Helicobacter pylori
infection? Meta-analysis of ran-
domized controlled trials. Intern
Med 2010;49:1103–1109.
4. Graham DY, Dore MP. Helicobacter
pylori therapy: a paradigm shift.
Expert Rev Anti Infect Ther 2016
May 3 [Epub ahead of print].
5. Graham DY, Lee YC, Wu MS.
Rational Helicobacter pylori ther-
apy: evidence-based medicine
rather than medicine-based evi-
dence. Clin Gastroenterol Hepatol
2014;12:177–186.
6. Liou JM, Chen CC, Chen MJ, et al.
Sequential versus triple therapy for
the first-line treatment of Heli-
cobacter pylori: a multicentre,
open-label, randomised trial. Lan-
cet 2013;381:205–213.
7. Wu JY, Liou JM, Graham DY. Evi-
dence-based recommendations for
successful Helicobacter pylori
11
COMMENTARIES
treatment. Expert Rev Gastro-
enterol Hepatol 2014;8:21–28.
8. Graham DY. Helicobacter pylori up-
date: Gastric cancer, reliable therapy,
and possible benefits. Gastroenter-
ology 2015;148:719–731.
9. Cosme A, Montes M, Martos M,
et al. Usefulness of antimicrobial
susceptibility in the eradication of
Helicobacter pylori. Clin Microbiol
Infect 2013;19:379–383.
10. Koletzko S, Jones NL, Goodman KJ,
et al. Evidence-based guidelines
from ESPGHAN and NASPGHAN
for Helicobacter pylori infection in
children. J Pediatr Gastroenterol
Nutr 2011;53:230–243.
11. Gold BD. Resistance testing for
Helicobacter pylori infection: is it
finally ready for prime time? J Pediatr
Gastroenterol Nutr 2014;59:3–4.
12. Shiota S, Reddy R, Alsarraj A, et al.
Antibiotic resistance of Heli-
cobacter pylori among male United
States veterans. Clin Gastroenterol
Hepatol 2015;13:1616–1624.
13. Lu H, Zhang W, Graham DY.
Bismuth-containing quadruple ther-
apy for Helicobacter pylori: lessons
from China. Eur J Gastroenterol
Hepatol 2013;25:1134–1140.
12
14. Graham DY, Lee SY. How to
effectively use bismuth quadruple
therapy: The good, the bad, and
the ugly. Gastroenterol Clin North
Am 2015;44:537–563.
15. Zhang W, Chen Q, Liang X, et al.
Bismuth, lansoprazole, amoxicillin
and metronidazole or clari-
thromycin as first-line Helicobacter
pylori therapy. Gut 2015;64:
1715–1720.
16. Dore MP, Lu H, Graham DY. Role
of bismuth in improving Heli-
cobacter pylori eradication with tri-
ple therapy. Gut 2016;65:870–878.
17. Megraud F, Coenen S,
Versporten A, et al. Helicobacter
pylori resistance to antibiotics in
Europe and its relationship to anti-
biotic consumption. Gut 2013;
62:34–42.
18. Ghotaslou R, Leylabadlo HE,
Asl YM. Prevalence of antibiotic
resistance in Helicobacter pylori: A
recent literature review. World J
Methodol 2015;5:164–174.
19. Thung I, Aramin H, Vavinskaya V,
et al. Review article: the global
emergence of Helicobacter pylori
antibiotic resistance. Aliment Phar-
macol Ther 2016;43:514–533.
20. Duck WM, Sobel J, Pruckler JM,
et al. Antimicrobial resistance inci-
dence and risk factors among Hel-
icobacter pylori-infected persons,
United States. Emerg Infect Dis
2004;10:1088–1094.
21. Scott D, Weeks D, Melchers K,
et al. The life and death of Heli-
cobacter pylori. Gut 1998;43 Suppl
1:S56–S60.
Conflicts of interest
The authors have made the following disclosures:
Dr Graham is a consultant for RedHill Biopharma
regarding novel Helicobacter pylori therapies and
has received research support for culture of H
pylori and is the principal investigator of an
international study of the use of antimycobacterial
therapy for Crohn’s disease. He is also a
consultant for BioGaia in relation to probiotic
therapy for H pylori infection. Dr Laine has
nothing to declare.
Funding
Dr Graham is supported in part by the Office of
Research and Development Medical Research
Service Department of Veterans Affairs (VA), and
National Institutes of Health (NIH) grant DK56338
which supports the Texas Medical Center
Digestive Diseases Center. The contents are
solely the responsibility of the authors and do not
necessarily represent the official views of the VA
or NIH.
Most current article
© 2016 by the AGA Institute
0016-5085/$36.00
http://dx.doi.org/10.1053/j.gastro.2016.05.009