evaluations regarding the GRADE approach for the statements
were also reviewed, and all panelists discussed the
findings and
other issues before
finalization of the phrasing for individual
statements. Any PICO components that are unequivocally
implied were removed from the
final statements to make the
message clearer to the readers. Finally, participants were asked
to vote on their level of agreement for each speci
fic statement.
A statement was accepted if
>75% of participants voted 4
(agree) or 5 (strongly agree) on a scale of 1 to 5 (with 1, 2, and
3 representing disagree strongly, disagree, and uncertain,
respectively).
Once a statement was accepted, the participants then voted
on the
“strength” of the recommendation, which was accepted
with a 51% vote. Per the GRADE system, the strength of each
recommendation was assigned as strong (
“we recommend.”)
or conditional (
“we suggest.”). The strength of the recom-
mendation considers risk-bene
fit balance, patients’ values and
preferences, cost and resource allocation, and quality of the
evidence. Therefore, it is possible for a recommendation to be
classi
fied as strong despite having low-quality evidence to
support it or conditional despite the existence of high-quality
evidence to support it.
39
Based on the GRADE approach, a
strong recommendation indicates the statement should be
applied in most cases, while a conditional recommendation
signi
fies that clinicians “.should recognize that different
choices will be appropriate for different patients and that they
must help each patient to arrive at a management decision
consistent with her or his values and preferences.
”
39
The steering committee drafted the initial manuscript, which
was revised by all members of the consensus group and all au-
thors, after which it was made available to all members of CAG
for comments before submission for publication. Per CAG policy,
all participants provided written disclosure of relevant potential
con
flicts of interest for the 24 months before the meeting, which
were made available to the other group members.
Role of the Funding Sources
CAG administered all aspects of the meeting, which was
cofunded by CAG and the Canadian Helicobacter Study Group
with no external funding sources.
Recommendation Statements
The individual recommendation statements are pro-
vided and include the quality of supporting evidence as
assessed by the GRADE method and the voting results; a
discussion of the evidence considered for the speci
fic
statement is also presented. The quality of evidence was
determined to be low for some statements, largely because
of high risk of bias (most often due to lack of adequate
blinding). Acknowledging the importance of quality of
evidence, the consensus group also considered other fac-
tors in issuing strong rather than conditional recommen-
dations for certain statements despite lower quality of
evidence. The strength of these recommendations was
driven by consequences of therapeutic failure, including
the negative consequences of peptic ulcer disease, such as
gastrointestinal bleeding, an increased risk of the devel-
opment of gastric cancers, and an increased risk of the
development of resistant strains.
20,40,41
In addition, erad-
ication success is highest with initial therapy and
decreases with subsequent rescue therapy attempts.
42,43
Hence, a treatment option may have been strongly rec-
ommended even if the evidence was not high quality to
avoid the negative consequences of failure.
A summary of the recommendation statements is
provided in
Table 3
. The most important evidence for
each of the statements is summarized in
Supplementary
Tables 1 to 14
.
All Patients
Statement 1. In patients with
H pylori infection,
we recommend a treatment duration of 14 days.
GRADE: Strong recommendation; quality of evidence moder-
ate for PAC and very low for PBMT, PAMC, and PAL. Vote:
strongly agree, 87.5%; agree, 12.5%.
Key evidence (
Supplementary Table 1
). A Cochrane
meta-analysis of RCTs found that a 14-day duration of PPI
triple therapy was associated with a signi
ficantly greater
proportion of eradication compared with shorter durations
(ITT: 45 studies, 14 vs 7 days, 82% vs 73%; 12 studies, 14
vs 10 days, 84% vs 79%).
28
A signi
ficant effect was seen in
54
Fallone et al
Gastroenterology Vol. 151, No. 1
the PAC subgroup (34 studies of 14 vs 7 days; relative risk
of H pylori persistence, 0.65 [95% con
fidence interval {CI},
0.57
–0.75]; number needed to treat [NNT], 12 [95% CI,
9
–16]) as well as in the PPI, amoxicillin, and quinolone
subgroup (2 studies of 14 vs 7 days; relative risk, 0.37 [95%
CI, 0.16
–0.83]; NNT, 3 [95% CI, 2–10]) (
Table 4
). There was
Table 3.Summary of Consensus Recommendations for the Treatment of H pylori Infection
All patients
1. In patients with H pylori infection, we recommend a treatment duration of 14 days. GRADE: Strong recommendation; quality of evidence
moderate for PAC and very low for PBMT, PAMC, and PAL.
First-line therapy
2. In patients with H pylori infection, we recommend that the choice of
first-line therapy consider regional antibiotic resistance patterns and
eradication rates. GRADE: Strong recommendation; quality of evidence low.
3. In patients with H pylori infection, we recommend traditional bismuth quadruple therapy (PBMT) for 14 days as one of the options for
first-line therapy. GRADE: Strong recommendation; quality of evidence moderate for efficacy and very low for duration.
4. In patients with H pylori infection, we recommend concomitant nonbismuth quadruple therapy (PAMC) for 14 days as one of the options
for
first-line therapy. GRADE: Strong recommendation; quality of evidence moderate for efficacy and very low for duration.
5. In patients with H pylori infection, we recommend restricting the use of PPI triple therapy (PAC or PMC for 14 days) to areas with known low
clarithromycin resistance (
<15%) or proven high local eradication rates (>85%). GRADE: Strong recommendation; quality of evidence
moderate for ef
ficacy of PPI triple therapy for 14 days and low for restrictions.
6. In patients with H pylori infection, we recommend against the use of levo
floxacin triple therapy (PAL) as a first-line therapy. GRADE: Strong
recommendation; quality of evidence very low.
7. In patients with H pylori infection, we recommend against the use of sequential nonbismuth quadruple therapy (PA followed by PMC) as
a
first-line therapy. GRADE: Strong recommendation; quality of evidence moderate.
Prior failure
8. In patients who have previously failed to respond to H pylori eradication therapy, we recommend traditional bismuth quadruple therapy
(PBMT) for 14 days as an option for subsequent therapy. GRADE: Strong recommendation; quality of evidence low.
9. In patients who have previously failed to respond to H pylori eradication therapy, we suggest levo
floxacin-containing therapy for 14 days
as an option for subsequent therapy. GRADE: Conditional recommendation; quality of evidence low.
10. In patients who have previously failed to respond a clarithromycin-containing H pylori eradication therapy, we recommend against the
use of clarithromycin-containing regimens as subsequent therapy. GRADE: Strong recommendation; quality of evidence low.
11. In patients who have previously failed to respond to a levo
floxacin-containing H pylori eradication therapy, we recommend against the
use of levo
floxacin-containing regimens as subsequent therapy. GRADE: Strong recommendation; quality of evidence low.
12. In patients who have previously failed to respond to H pylori eradication therapy, we recommend against the use of sequential nonbismuth
quadruple therapy (PA followed by PMC) as an option for subsequent therapy. GRADE: Strong recommendation; quality of evidence
very low.
13. We recommend restricting the use of rifabutin-containing regimens to cases in which at least 3 recommended options have failed.
GRADE: Strong recommendation; quality of evidence very low.
Supplemental therapy
14. In patients with H pylori infection, we recommend against routinely adding probiotics to eradication therapy for the purpose of
reducing adverse events. GRADE: Strong recommendation; quality of evidence very low.
15. In patients with H pylori infection, we recommend against adding probiotics to eradication therapy for the purpose of increasing
eradication rates. GRADE: Strong recommendation; quality of evidence very low.
NOTE. The consensus group concluded that there was insuf
ficient evidence to support or refute the efficacy of PAMC as a
second-line option and thus was unable to recommend for or against this regimen as a rescue therapy. Similarly, the group
concluded that there was insuf
ficient evidence to make a recommendation on high-dose dual therapy with a PPI and
amoxicillin. See
Tables 1
and
2
for more details on regimens and dosing.
Table 4.Relative risks for H pylori Persistence According to Duration of Regimen
Studies (n
¼ 75)
14 vs 7 days
10 vs 7 days
14 vs 10 days
PPI triple therapy (n
¼ 59)
0.66 (0.60
–0.74);
NNT, 11 (9
–14); (n ¼ 45)
0.80 (0.72
–0.89);
NNT, 21 (15
–38); (n ¼ 24)
0.72 (0.58
–0.90);
NNT, 17 (11
–46); (n ¼ 12)
PAC (n
¼ 34)
0.65 (0.57
–0.75);
NNT, 12 (9
–16); (n ¼ 34)
0.80 (0.70
–0.91);
NNT, 21 (14
–48); (n ¼ 17)
0.69 (0.52
–0.91);
NNT, 16 (10
–54); (n ¼ 10)
PMC (n
¼ 4)
0.87 (0.71
–1.07); (n ¼ 4)
0.99 (0.55
–1.79); (n ¼ 2)
—
PAQ (n
¼ 2)
0.37 (0.16
–0.83);
NNT, 3 (2
–10); (n ¼ 2)
0.58 (0.36
–0.95);
NNT, 7 (5
–59); (n ¼ 2)
—
PPI bismuth quadruple
therapy (n
¼ 6)
0.71 (0.44
–1.15); (n ¼ 3)
0.70 (0.43
–1.14); (n ¼ 2)
1.13 (0.59
–2.18); (n ¼ 1)
NOTE. Based on data from a meta-analysis by Yuan et al.
28
Values are relative risk for H pylori persistence (95% CI); NNT (95%
CI); studies (n).
PAQ, PPI
þ amoxicillin þ quinolone.
July 2016
Toronto Consensus for
H pylori Treatment
55
no increase in discontinuations due to adverse events with
increasing duration of therapy.
With regard to quadruple therapies, a systematic review
of cohort studies found a trend toward greater treatment
success with longer durations (from 3 to 10 days) of non-
bismuth quadruple therapy (PAMC) (see statement 4).
44
A
14-day optimized PAMC combination also achieved higher
eradication rates compared with standard 10-day PAMC
(ITT, 93% vs. 87%; P
< .01); however, the optimized
regimen was not only of longer duration but also included
an increased PPI dose.
45
Finally, for bismuth quadruple therapy, the Cochrane
meta-analysis of RCTs did not
find that duration has a sig-
ni
ficant effect on therapeutic success for first-line therapy
(
Table 4
),
28
but there are very few studies with this com-
parison and a trend was suggested (see statement 3). A
meta-analysis
performed
for
the
consensus
meeting
assessed the duration of this regimen for the treatment of
those who previously failed to achieve eradication. Overall,
51 RCT and cohort studies were included (see statement 8),
and meta-analysis showed that the ITT eradication rate
was numerically but not statistically higher with the
14-day regimen versus the 10-day regimen (78.7% vs
75.6%; P
¼ .33).
Other issues and discussion. The increasing prevalence
of resistant strains of H pylori has led to increasing pro-
portions of failure of traditional H pylori treatments.
18
–22
In
a RCT of clarithromycin-containing triple therapies, the
eradication success rate of resistant strains was 35% lower
than that of sensitive strains.
46
The impact was greatest
among regimens of the shortest duration; the eradication
success rate of sensitive versus resistant strains was 42%
higher in the 7-day group, 33% higher in the 10-day group,
and 22% higher in the 14-day group. Therefore, indirect
evidence supports increased ef
ficacy with longer durations
of therapy in resistant strains. The differences in ef
ficacy
between therapies in the studies presented are likely
underestimated, because many of the studies are older and
the proportion of resistant strains has increased since they
were conducted.
Decisions. In light of the higher eradication rates with
longer durations of therapy compared with regimens of
shorter durations, the consensus group strongly recom-
mended that all H pylori regimens (both
first-line and rescue
therapies) be administered for 14 days. This prolonged use
of antibiotics for all patients is warranted because the
increased failures with shorter regimens would result in
resistant strains and less successful future treatments. It is
best to achieve the maximum cure rates from the start.
First-Line Therapy
Statement 2. In patients with
H pylori infection,
we recommend that the choice of
first-line therapy
consider regional antibiotic resistance patterns and
eradication rates. GRADE: Strong recommendation; qual-
ity of evidence low. Vote: strongly agree, 100%.
Key evidence (
Supplementary Table 2
). Although no
study directly examined the impact of tailoring
first-line
therapy
to
local
antibiotic
resistance
patterns
and
eradication rates, a meta-analysis of 5 RCTs (n
¼ 701) found
that culture-guided triple therapy resulted in a signi
ficantly
lower risk of treatment failure compared with empirical
standard triple therapy (ITT relative risk, 0.84; 95% CI,
0.77
–0.90; P < .00001; eradication rate, 85.4% vs 71.5%).
47
Other issues and discussion. H pylori
–resistant strains
have become more prevalent over time.
19
Studies from the
1990s showed a low prevalence of clarithromycin resistance
ranging from 1% to 8%,
18,19
which has risen to 16% to 24%
in more recent studies from around the world.
48
–50
Primary
resistance to metronidazole appears to have remained
relatively stable over time at 20% to 40%.
4,18
–20,51
H pylori
resistance to amoxicillin generally remains low at approxi-
mately 1% to 3%.
4,18,19,50,51
In addition, the prevalence of secondary resistance to
clarithromycin and metronidazole is very high: up to 67% to
82%
for
clarithromycin
and
52%
to
77%
for
metronidazole.
20,40,50
RCTs con
firm that the proportion of successful eradi-
cation is signi
ficantly lower in resistant compared with
sensitive strains, especially with triple therapy
46,52
–55
and
therapy of shorter duration.
46
The increasing prevalence of
clarithromycin
resistance
is
likely
the
main
factor
contributing to the increasing failure of non
–culture-guided
H pylori therapies over time, especially clarithromycin-
based triple therapies.
21,22
A meta-analysis of 12 studies
found that success of eradication with bismuth quadruple
therapy remained stable at approximately 80% in studies
from 2006 to 2011 compared with those from 2000 to
2005, but the ef
ficacy of clarithromycin-based triple ther-
apy decreased from approximately 80% in studies from
2000 to 2005 to only 62% in more recent studies
(2006
–2011) (
Figure 1
).
22
Bismuth quadruple therapy is unaffected by clari-
thromycin resistance.
22,53
However, the eradication success
Figure 1. Pooled successful eradication (ITT) in subgroup
analysis according to year of study publication. Based on
data from a meta-analysis by Venerito et al.
22
RDs are shown
as proportions rather than percentages.
56
Fallone et al
Gastroenterology Vol. 151, No. 1
rate with PBMT seems to be slightly lower in metronidazole-
resistant versus metronidazole-sensitive strains (92% vs.
80%; P
¼ .06).
53,54
In one meta-analysis of triple and
quadruple regimens, the successful eradication rate was
found to decrease by 0.5% for every 1% increase in the
prevalence of metronidazole resistance, suggesting that
when metronidazole resistance is 30%, treatment ef
ficacy
decreases by 15%.
56
Similar effects of resistance have been seen with levo-
floxacin triple therapy and bismuth quadruple levofloxacin-
based therapy; among levo
floxacin-susceptible strains, the
eradication rate was 97% in both groups; however, among
resistant strains, the proportion dropped to 71% with
quadruple therapy and 38% with triple therapy.
52
If the susceptibility pro
file of a patient’s infection or an
estimate of it from the patient
’s population is known, the
ef
ficacy of a proposed regimen can be predicted.
57
–59
Un-
fortunately, the resistance data required for these pre-
dictions is not available in most areas. Pragmatically, a
combination of local experience of treatment success with
different regimens and the patient
’s pretreatment exposure
to antibiotics can also aid in the identi
fication of the regimen
most likely to succeed.
60
Decisions. Evidence suggests that culture-guided therapy
is associated with higher eradication success rates
47
and that
both antibiotic-resistant H pylori
18,19,48
–50
and treatment
failures
46,52
–55
are increasing. Therefore, it is important to
encourage susceptibility testing to be made available locally
and performed if the patient is undergoing endoscopy.
However, it is not currently clinically practical or often
possible to perform susceptibility testing in all patients.
Therefore, the consensus group advised that local suscepti-
bility patterns be used as a helpful surrogate when available.
Studies to determine the local prevalence of primary antibi-
otic resistance patterns are essential to assist clinicians in
selecting the most appropriate
first-line treatment for their
practice. When available, the actual proportion of patients
with successful eradication after receiving a speci
fic treat-
ment can be used to guide future treatment selection. As
such, clinicians are encouraged to maintain records of the
eradication rates they obtain locally with treatments.
Statement 3. In patients with
H pylori infection,
we recommend traditional bismuth quadruple ther-
apy (PBMT) for 14 days as one of the options for
first-line therapy. GRADE: Strong recommendation; qual-
ity of evidence moderate for ef
ficacy and very low for dura-
tion. Vote: strongly agree, 75%; agree, 25%.
Key evidence (
Supplementary Table 3
). Two systematic
reviews of RCTs have evaluated the ef
ficacy of first-line
bismuth quadruple therapy (PBMT) compared with triple
therapy (PAC).
21,22
The more recent meta-analysis of 12
RCTs found that the overall pooled eradication success rate
was 77.6% with PBMT and 68.9% with PAC (risk difference
[RD], 6%; 95% CI,
À1% to 13%; note that the mathematical
difference in the eradication success rate is not the same as
the RD because the latter statistic is more appropriately
weighted for the study effect size and precision of each
estimate).
22
Although this analysis did not show a statisti-
cally signi
ficant difference, there was a trend toward greater
eradication rates with PBMT.
22
The subgroup analysis of
duration showed that 10-day quadruple therapy was more
effective than 7-day triple therapy, but no differences were
noted between the therapies when given for the same
duration for either 7 days or 10 to 14 days. Speci
fic analyses
for 14-day PBMT were not performed. Only one study was
found that directly compared 14-day durations in
first-line
therapy, which showed higher eradication success rates
with bismuth quadruple therapy compared with triple
therapy; however, this was signi
ficant only in the per-
protocol analysis and not the ITT analysis.
61
In addition,
antimicrobial resistance has been shown to have less impact
on the success of PBMT regimens (metronidazole-sensitive
vs -resistant strains, 89.4% vs 80.6%) compared with PAC
regimens (clarithromycin-sensitive vs -resistant strains,
90.2% vs 22.2%).
21
Other issues and discussion. As described in statements
1 and 8, a meta-analysis of observational data conducted for
the meeting to evaluate the duration of bismuth quadruple Dostları ilə paylaş: |