original article
Wien Klin Wochenschr (2010) 122: 413–422
DOI 10.1007/s00508-010-1404-3
© Springer-Verlag 2010
Printed in Austria
Advantages of Moxifloxacin and Levofloxacin-based
triple therapy for second-line treatments of persistent
Helicobacter pylori infection: a meta analysis
Yuqin Li*, Xiayue Huang*, Linhua Yao, Ruihua Shi, Guoxin Zhang
Department of Gastroenterology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
Received November 2, 2009, accepted after revision May 14, 2010, published online July 16, 2010
Vorteile einer Moxifloxazin beziehungsweise
Laevofloxazin basierten Triple Therapie als
Second-line Behandlung einer persistenten
Infektion mit Helicobacter pylori: Eine
Metaanalyse
Zusammenfassung. Ziel: Das wesentliche Ziel der vorlie-
genden Meta Analyse war es, die Wirksamkeit und Sicher-
heit einer Therapie einer persistierenden Helicobacter pylori
Infektion mit entweder einer Clarithromycin und 2. Gene-
ration Fluorquinolon-basierten Tripel Therapie mit einer
Bismuth basierten Quadrupel Therapie zu vergleichen.
Methodik: Es wurde eine systematische Literatur
Recherche nach Artikel und Abstracts des Zeitraums 1981–
2009 durchgeführt. Durchforstet wurden Medline, Pub-
Med, EMBase, Google Scholar sowie CNKI (chinesisch),
Wanfang (chinesisch) Digital Database und recent Diges-
tive Disease Week, United European Gastroenterology
Week sowie Konferenzen der European Helicobacter Study
Group. Die stufenweise Einengung oder Erweiterung der
Recherche erfolgte durch Boolean operators (NOT, AND,
OR). 16 Artikel und 4 Abstracts erfüllten die Einschlusskri-
terien und wurden in die Meta Analyse (Review Manager
4.2.8) einbezogen.
Ergebnisse: Die berichteten Eradikationsraten zeigten,
dass die Clarithromycin basierte Triple Therapie der Bis-
muth basierten Quadrupel Therapie unterlegen zu sein
scheint (OR = 0,53; 95 % CI: 0,35–0,80; P = 0,002). 13 RCTs
verglichen eine Laevofloxacin basierte Triple Therapie mit
einer Bismuth basierten Quadrupel Therapie – diese 2
Therapiearten unterschieden sich bezüglich ihres Eradi-
kationserfolges nicht signifikant (OR = 1,43; 95 % CI: 0,82–
2,51; P = 0,21). Allerdings waren die Eradikationsraten der
* These authors have contributed equally to this paper.
Correspondence: Prof. Guoxin Zhang, Department of
Gastroenterology, The First Affiliated Hospital of Nanjing Medical
University, 300 Guangzhou Road, Nanjing 210029, China,
E-mail: guoxinz@njmu.edu.cn
wkw 13–14/2010 © Springer-Verlag
Wiener klinische Wochenschrift
The Middle European Journal of Medicine
10-tägigen Laevofloxacin basierte Triple Therapie einer
7tägigen Bismuth basierten Quadrupel Therapie signifi-
kant überlegen (OR = 4,79; 95 % CI: 2,95–7,79; P < 0,00001).
Die Laevofloxacin basierte Tripel Therapie wurde besser
vertragen als die Bismuth basierte Quadrupel Therapie
(OR = 0,41; 95 % CI: 0,27–0,61; P < 0,0001) und musste auch
seltener wegen Nebenwirkungen abgebrochen werden
(OR = 0,13; 95 % CI: 0,06–0,33; P < 0,0001). Außerdem lässt
das Ergebnis unserer Meta Analyse vermuten, dass die
Eradiaktionsraten der Moxifloxacin-basierten Tripel
Therapie der Bismuth basierten Quadrupel Therapie
geringfügig – allerdings ohne statistische Signifikanz –
überlegen ist.
Schlussfolgerung: Eine 2. Generation Fluoroquinolon-
basierte Tripel Therapie – vor allem das 10tägige Regime
mit Laevofloxacin - kann als Behandlungsart 1. Wahl zur
Eradikation einer persistierenden Helicobacter pylori In-
fektion empfohlen werden.
Summary. Objective: The main aim of this meta-analysis
was to compare the efficacy and safety of clarithromycin
and second-generation fluoroquinolone-based triple ther-
apy vs. bismuth-based quadruple therapy for the treat-
ment of persistent Helicobacter pylori infection.
Methods: A systematic literature search was conducted
for articles and abstracts from 1981 to March 2009 using
Medline, PubMed, EMBase, Google Scholar and CNKI
(Chinese), Wanfang (Chinese) digital database and recent
Digestive Disease Week, United European Gastroenterol-
ogy Week, and European Helicobacter Study Group con-
ferences were also performed. Boolean operators (NOT,
AND, OR) were used in succession to narrow and widen
the search. Sixteen articles and four abstracts met the in-
clusion criteria, and were included in the meta-analysis by
using Review Manager 4.2.8.
Results: The eradication rates demonstrated that clari-
thromycin-based triple therapy is inferior to bismuth-
based quadruple therapy (OR = 0.53, 95% CI: 0.35–0.80,
P = 0.002). Thirteen RCTs compared levofloxacin-based
triple therapy vs. bismuth-based quadruple therapy, the
Second-line treatment for Helicobacter pylori infection 413
 original article
eradication rates of the two regimens were shown to have
no significant difference (OR = 1.43, 95% CI: 0.82–2.51,
P = 0.21). But the eradication rates demonstrated superior-
ity of the 10-day levofloxacin-based triple therapy over
7-day bismuth-based quadruple therapy (OR = 4.79, 95%
CI: 2.95–7.79, P < 0.00001). Levofloxacin-based triple ther-
apy was better tolerated than bismuth-based quadruple
therapy with lower rates of side effects (OR = 0.41, 95% CI:
0.27–0.61, P < 0.0001), and lower rates of discontinuation
of therapy due to adverse events (OR = 0.13, 95% CI: 0.06–
0.33, P < 0.0001). Furthermore, our meta-analysis sug-
gested that the eradication rates of the moxifloxacin-based
triple therapy has a slight superiority to bismuth-based
quadruple therapy, but there was no significant difference
between them.
Conclusion: Second-generation fluoroquinolone-based
triple therapy can be suggested as the regimen of choice for
rescue therapy in the eradication of persistent H. pylori in-
fection especially 10-day levofloxacin-based triple therapy.
Key words: PPI, bismuth, levofloxacin, moxifloxacin, fluo-
roquinolone, Helicobacter, eradication rate, second-line
treatment.
Introduction
Helicobacter pylori infection plays an important role in the
pathogenesis of chronic gastritis, peptic ulcer disease, gas-
tric mucosa-associated lymphoid tissue lymphoma, and
gastric malignancies [1–4]. Recently, several studies have
suggested that H. pylori is also involved in the develop-
ment of coronary heart disease [5], idiopathic thrombo-
penic purpura [6] and gastroesophageal reflux disease [7].
H. pylori is a major cause of illness and death worldwide
[4]; therefore, eradication of H. pylori has become an im-
portant consideration in the treatment of these diseases.
This is especially true regarding the natural history of both
peptic ulcer disease and gastric lymphoma [8]. Many
worldwide consensus conferences have recommended
triple therapy consisting of a proton-pump inhibitor (PPI)
and two antibiotics as first-line eradication therapy [9].
But, the increasing prevalence of H. pylori strains resistant
to antibiotics has led to increasing rates of treatment fail-
ure [10–12]. Thus, a number of second-line therapies are
recommended such as: quadruple therapy [proton-pump
inhibitor (PPI), bismuth, tetracycline and metronidazole]
for 1 week or 2 weeks, PPI-based triple therapies for 1 week
or 10 days or 2 weeks [13]. An ideal therapy should be
short, have few side effects, good compliance, the low rates
of bacterial resistance and high rates of eradication [14].
Although many studies have reported on the efficacy
PPI-based triple therapy vs. bismuth-based quadruple
therapy as second-line treatment of H. pylori infection,
there is no consensus. Some articles have reported that
bismuth-based quadruple therapy was preferred as the
second-line treatment of H. pylori infection because this
regimen had many advantages such as wide availability,
low cost and reasonably good efficacy. However, other ar-
ticles have shown that PPI-based triple therapies have
many advantages such as fewer side effects, better compli-
414 Second-line treatment for Helicobacter pylori infection
ance, lower resistance and higher eradication rates in favor
of this combination as the best regimen [14]. Many articles
have stated that the two therapies are equally effective as
second-line treatments for H. pylori infection. It is also
well known that in Europe, that the recommended treat-
ment duration is usually 7 days, whereas in the United
States, the Food and Drug Administration has approved
regimens of 7, 10, or 14 days [9, 15]. In fact, different treat-
ment duration may not significantly affect the results ac-
cording to many studies.
Recently, a meta-analysis of levofloxacin-based triple
therapy versus bismuth-based quadruple therapy for per-
sistent Helicobacter pylori infection was reported by Saad
et al. [14]. They showed that a 10-day course levofloxacin
triple therapy was more effective and better tolerated than
7-day bismuth-based quadruple therapy in the treatment
of persistent H. pylori infection. Moreover, another meta-
analysis of levofloxacin-based rescue regimens after Heli-
cobacter pylori treatment failure was reported by Gisbert
JP [16]. However, since the number of subjects in two me-
ta-analysis were not large enough, we added several prox-
imal studies to compare levofloxacin-based triple therapy
vs. bismuth-based quadruple therapy. In addition, they
only compared levofloxacin-based triple therapy with bis-
muth-based quadruple therapy. Moxifloxacin-based and
clarithromycin-based triple therapy, which are still ac-
cepted in many countries were not mentioned, and pa-
tients with drug resistance were also not shown. For this
main reason, we performed a systematic literature review
and meta-analysis of randomized controlled trials (RCTs)
with different regimens as salvage for Helicobacter pylori
infection. Our primary objective was to compare the
efficacy of fluoroquinolone-based triple therapy and bis-
muth-based quadruple therapy for the treatment of per-
sistent Helicobacter pylori infection. The second objective
was to compare the safety of the two therapies.
Materials and methods
Study sources and searches
We searched PubMeD, MEDLINE, EMBASE, Google Scholar,
CNKI (Chinese) and Wanfang (Chinese) digital database and re-
cent Digestive Disease Week, United European Gastroenterology
Week, and European Helicobacter Study Group conferences for
relevant articles and abstracts published in English and Chinese
from 1981 to March 2009. The search was limited to human stud-
ies, but was otherwise unrestricted. MeSH terms and keywords
used to identify articles included “second-line treatment”, “rescue
therapy”, “salvage therapy”, “Helicobacter pylori”, “H. pylori”, “pro-
ton-pump inhibitor”, “bismuth”, “levofloxacin”, “moxifloxacin”,
“quinolones”, “triple therapy”, “quadruple therapy”, and “treat-
ment”. Boolean operators (NOT, AND, OR) were used in succes-
sion to narrow and widen the search.
Study selection
For the meta-analysis, all studies had to meet the following inclu-
sion criteria: (1) A study had to describe an RCT. (2) All patients
had to have failed one foregoing course of H. pylori eradication
therapy. (3) Confirmation of infection eradication at least 4 weeks
after completion of treatment (based on urea breath test or gas-
© Springer-Verlag 13–14/2010 wkw

tric mucosal biopsy for histology or culture). (4) Either moxi-
floxacin and levofloxacin-based triple therapy, clarithromycin-
based triple therapy for 7, 10 and 14 days or bismuth-based
quadruple therapy for 7 and 14 days. (5) Abstracts or full articles
with complete data and written in English or Chinese.
Non-randomized studies were excluded, as were case reports,
letters, editorials, commentaries and reviews with insufficient de-
tails to meet the inclusion criteria.
Data extraction
Two independent reviewers extracted the data from each paper
fulfilling inclusion criteria to increase accuracy. If some studies
had discrepancies, a consensus through discussion was made.
For each study, the following data were examined: study design;
the number of patients in the study and in each treatment regi-
men; drug regimen, including treatment duration; methods used
to confirm eradication; number of patients was successfully erad-
icated; incidence of side effects; ratio of discontinuation due to
adverse events and drug resistance. We used intention-to-treat
data analysis and assessed the quality of each study by using the
test of Jadad [17], since the data were not sufficient for PP analy-
sis (PP analysis: excluded patients with poor compliance of ther-
apy and patients with unevaluable data after therapy).
Statistical analysis
The analysis was performed in Review Manager 4.2.8. The odds
ratio (OR) and 95% confidence interval (CI) for H. pylori eradica-
tion rates were estimated for each study in a random-effects
model or in a fixed-effects model. For each section, we assessed
the heterogeneity of sub-study. Since tests of heterogeneity had a
relative low power, the threshold for P values was set at a high
level; a P value of <0.100 indicated significant heterogeneity. If
there was a significant heterogeneity (P < 0.100), we selected a
random-effects model to pool the data. If not, we selected a fixed-
effects model to pool the data. However, this test used in our re-
Articels from health literature search (n=183) abstracts initially searched ((n= 10)
Articles excluded (n=118) abstracts excluded (n =4)
Irrelevant to Helicobacter pylori:43 Irrelevant to second-line treatment: 79
Articles found to be relevant to H. pylori and second-line treatment (n= 65) and abstracts (n= 6)
Articles and abstracts excluded (n=51)
RCTs comparing one drug (n=8) and in abstracts (n=2)
Articles without RCTs or incomplete data: (n=41)
Articles that met inclusion criteria (n=16) abstracts (n =4)
Fig. 1. Identification of eligible randomized, controlled trials (RCTs)
wkw 13–14/2010 © Springer-Verlag
original article
port had limitations. The I2 statistic was further used to estimate
statistical heterogeneity among all of the studies. An I2 value of
30% or greater indicated substantial heterogeneity.
Results
Characteristics of the articles in our meta-analysis
Overall, 16 articles out of 183 relevant reports and 4 ab-
stracts were identified to meet the inclusion criteria, and
were included in the meta-analysis [18–37]. Of these articles
and abstracts, 17 were published in English [18–37] and
three [20–22] in Chinese. The characteristics of eligible ran-
domized controlled trials (RCTs) are summarized in Fig. 1.
After review of the full-text articles, 118 articles and 4
abstracts were excluded because they were irrelevant to
H. pylori or irrelevant to second-line treatment. An
additional 41 RCTs were excluded because they did not
meet our inclusion criteria. Another eight articles and two
abstracts were also excluded because these studies only
tested the efficacy of a single drug between different thera-
pies, which precluded inclusion in the meta-analysis.
Table 1 shows information on the RCTs. Thirteen RCTs com-
pared levofloxacin-based triple therapy with bismuth-based
quadruple therapy. Three RCTs compared levofloxacin-
based triple therapy with bismuth-based quadruple therapy
for patients with clarithromycin and metronidazole-resist-
ant strains. Three RCTs compared moxifloxacin-based triple
therapy with bismuth-based quadruple therapy. Three
RCTs compared clarithromycin-based triple therapy with
bismuth-based quadruple therapy. Two additional pro-
spective RCTs met inclusion criteria for the pooled analysis
assessing effects of duration of therapy on eradication rates.
All articles reported results as intention-to-treat analyses.
Second-line treatment for Helicobacter pylori infection 415
 original article

Table 1. Summary of articles included in the meta-analysis

References Treatment Eradication rate, Eradication rate in patients Side effects Discontinuation Eradication rate of H.
n/N(N′) with clarithromycin of treatment rate due to pylori with metronidazole
resistant H. pylori adverse events in resistance
Kuo CH 2009 EaAL7 58/83(77) NR 10/83(77) 1/83(77) NR
[18] EaMBT7 53/83(63) NR 25/83(63) 5/83(63) NR
Jung HS 2008 PaAL7 16/31(30) NR 3/31(30) 0/31(30) NR
[19] PaMBT7 22/45(35) NR 1145(35)/ 1/45(35) NR
Zhang H 2008 EaAL7 42/48 NR 7/48 0/48 NR
[20] EaMBT7 33/47 NR 15/47 5/47 NR
Liang ZG 2007 EaAL10 33/38 NR 2/38 NR NR
[21] EafuBM14 22/34 NR 10/34 NR NR
Zhang Y 2007 EaAL7 42/49(45) NR 4/49(45) NR NR
[22] EaMBA7 30/44(39) NR 5/44(39) NR NR
Wong WM 2006 LaAL7 31/54(52) 21/32 18/54(52) NR 19/33
[23] LaMBT7 37/52(49) 14/21 21/52(49) NR 17/24
Nista EC 2005 RaAL7 37/50 NR NR 0/50 NR
abstract (01) [24] RaAL10 42/46 NR NR 0/50 NR
abstract (01) [24] RaMBT7 34/50 NR NR 4/50 NR
Wong WM 2004 LaAL7 21/33 NR NR NR NR
abstract (02) [25] LaMBT7 22/30 NR NR NR NR
Nista EC 2004 EaAL10 26/30 NR 8/30 0/30 NR
abstract (03) [26] EaMBT7 25/35 NR 21/35 4/35 NR
Bilard C 2004 PaAL10 31/44(41) 3/5 11/44(41) 1/44(41) 12/19
[27] OaMBT7 17/46(41) 2/6 13/46(41) 1/46(41) 5/22
Wong WM 2003 RaRL7 51/56(54) 24/28 19/56(54) NR 29/33
[28] RaMBT7 48/53(52) 24/26 31/53(52) NR 18/21
Perri F 2003 PaAL7 38/60(58) NR 3/60(58) 1/60(58) NR
[29] PaMBT7 50/60(55) NR 17/60(55) 3/60(55) NR
Nista EC 2003 RaAL10 66/70(70) NR 10/70(70) 0/70 NR
[30] RaTnL10 63/70(70) NR 11/70(70) 0/70 NR
[30] RaMBT7 44/70(64) NR 21/70(64) 5/70(64) NR
[30] RaMBT14 48/70(64) NR 33/70(64) 7/70(64) NR
Kang JM 2007 EaM″A10 100/139(121) NR 17/139(121) 1/139(121) NR
[31] EaMBT14 38/53(42) NR 21/53(42) 7/53(42) NR
Cheon JH 2006 EaM″A7 31/41(37) NR 4/41(37) 1/41(37) NR
[32] EaMBT7 24/44(33) NR 12/44(33) 4/44(33) NR
Bago J 2009 Oa M″M 60/82(76) NR 12/82(76) 4/82(76) NR
[33] OaMBT 42/78(65) NR 18/78(65) 11/78(65) NR
Peitz U 2002 OaAC7 19/44(38) 7/23 15/23 6/44(38) 13/29
[34] OaMBT7 27/40(39) 22/34 15/34 1/40(39) 25/40
Qasim A 2005 PPI+A+C 40/87 NR NR NR NR
[35] PPI+B+M+T 112/183 NR NR NR NR
Peitz U 1998 OaAC7 15/28 6/15 NR NR 11/21
abstract (04) [36] OaMBT7 17/29 10/17 NR NR 12/21
Quadruple therapy
(7days vs. 14 days)
Nista EC 2003 RaMBT7 44/70(64) / 21/70(64) 5/70(64) /
[30] RaMBT14 48/70(64) / 33/70(64) 7/70(64) /
Mantzaris GJ 2005 OaBMT7 36/54(45) / 7/54(45) 3/54(45) /
[37] OaBMT14 48/61(50) / 14/61(500 7/61(50) /
Oa omeprazole; La lansoprazole; Ra rabeprazole; Ea esomeprazole; Pa pantoprazole; A amoxicillin; C clarithromyc; M metronidazole; L levofloxacin; B bismuth;
M ″ moxifloxacin; R rifabutin; T tetracycline; Tn tinidazole; NR not reported; n the number of patients who were successfully eradicated; N the total number of
patients who received eradication; N ′ the total number of patients who subduced patients with poor compliance of therapy and unevaluable data after therapy.

416 Second-line treatment for Helicobacter pylori infection © Springer-Verlag 13–14/2010 wkw

Effect of PPI-based triple therapies (clarithromycin-based
triple therapy and levofloxacin-based triple therapy
and moxifloxacin-based triple therapy) vs. bismuth-based
quadruple therapy for second-line treatment
of Helicobacter pylori infection
We divided all PPI-based triple therapies into several
groups, and then compared them with bismuth-based
quadruple therapy. The eradication rates showed signifi-
cant difference between triple therapies based on clari-
thromycin and bismuth-based quadruple therapy. The
eradication rates with intention-to-treat analyses demon-
strated that clarithromycin-based triple therapy is inferior
to bismuth-based quadruple therapy. The results for two
regimens were as follows: clarithromycin-based: 46.54%,
bismuth-based: 61.90% (Fig. 2, OR = 0.53, 95% CI: 0.35–
0.80, P = 0.002). The test for heterogeneity in Fig. 2 showed
no significant difference between articles (P = 0.52, I2 = 0%).
Besides that, we compared second-generation fluoroqui-
nolone-based triple therapies (which included both levo-
floxacin-based triple therapy and moxifloxacin-based
triple therapy) with bismuth-based quadruple therapy.
Although the eradication rates with intention-to-treat
analyses demonstrated superiority of the moxifloxacin-
based triple therapy to bismuth-based quadruple therapy
(Fig. 3, OR = 1.78, 95% CI: 0.98–3.22, P = 0.06), there was no
Fig. 2. The eradication rates with intention-to treat analyses demonstrated that clarithromycin-based triple therapy is inferior to bismuth-based
quadruple therapy. No heterogeneity among studies
Fig. 3. The eradication rates with intention-to treat analyses demonstrated a slightly superiority of the moxifloxacin-based triple therapy
to bismuth-based quadruple therapy, but there was no significant difference between them. Heterogeneity among studies
wkw 13–14/2010 © Springer-Verlag
original article
significant difference between them. The test for heteroge-
neity in Fig. 3 showed significant difference between arti-
cles. Additional analyses between levofloxacin-based
triple therapy and bismuth-based quadruple therapy re-
vealed that there was no significant difference between
them. The odds ratio (OR) and 95% confidence interval
(CI) for them were OR = 1.43, 95% CI: 0.82–2.51, P = 0.21
(figure not shown).
7-day and 10-day Levofloxacin-based triple therapy
vs. 7-day Bismuth-based quadruple therapy
for persistent H. pylori
Analysis of nine RCTs was shown to have no significant dif-
ference (Fig. 4, OR = 1.04, 95% CI: 0.78–1.39, P = 0.81) be-
tween 7-day levofloxacin-based triple therapy and 7-day
bismuth-based quadruple therapy. The test for heteroge-
neity in Fig. 4 showed significant difference between arti-
cles (P = 0.02, I2 = 55.1%). Moreover, there were four RCTs
compared 10-day levofloxacin-based triple therapy vs.
7-day bismuth-based quadruple therapy, the eradication
rates demonstrated superiority of the 10-day levofloxacin-
based triple therapy over 7-day bismuth-based quadruple
therapy (Fig. 5, OR = 4.79, 95% CI: 2.95–7.79, P < 0.00001).
The test for heterogeneity in Fig. 5 showed no significant
difference between articles (P = 0.60, I2 = 0%).
Second-line treatment for Helicobacter pylori infection 417
 original article
Fig. 4. The eradication rates of 7-day levofloxacin-based triple therapy and 7-day bismuth-based quadruple therapy. Heterogeneity among studies
Fig. 5. The eradication rates of 10-day levofloxacin-based triple therapy and 7-day bismuth-based quadruple therapy. No heterogeneity among studies
The frequency of side effects of treatment, ratio
of discontinuation of therapies due to adverse events
between levofloxacin-based triple therapy
and bismuth-based quadruple therapy
There were eleven RCTs that reported side effects of treat-
ment. The frequency of side effects with levofloxacin-
based triple therapy and bismuth-based quadruple salvage
therapy was 16.75% and 32.71%, respectively. Our analysis
suggested that the frequency of side effects of levofloxacin-
based triple therapy was less common than bismuth-based
quadruple therapy (Fig. 6, OR = 0.41, 95% CI: 0.27–0.61,
P < 0.0001). Eight RCTs revealed that the rate of
discontinuation of therapy due to adverse events for
levofloxacin-based triple therapy compared to bismuth-
based quadruple salvage therapy were 0.56% and 6.92%,
respectively. The results also showed that levofloxacin-
based triple therapy had lower rates of discontinuation of
therapy due to adverse events (Fig. 7, OR = 0.13, 95% CI:
0.06–0.33, P < 0.0001). The tests for heterogeneity in Fig. 6
showed significant difference between articles and Fig. 7
showed no significant difference between articles.
418 Second-line treatment for Helicobacter pylori infection
Levofloxacin-based triple therapy vs. bismuth-based
quadruple therapy for persistent H. pylori with drug
resistance
In three RCTs of patients with clarithromycin and metroni-
dazole-resistant strains, eradication of clarithromycin and
metronidazole-resistant strains was shown to have no sig-
nificant difference for levofloxacin-based triple therapy and
bismuth-based quadruple therapy (figure not shown).
Different duration of treatment between quadruple
therapy
Furthermore, we determined whether the eradication rate
was different in quadruple therapy with different duration
of treatment, and found there was no significant difference
between them (OR = 0.67, 95% CI: 0.39–1.14, P = 0.14). But,
the sample size was too small to draw firm conclusions
from our analysis. Therefore, attempts to increase the
duration of quadruple therapy and prolong exposure to
antibiotics have not been demonstrated to result in mea-
surable benefit.
© Springer-Verlag 13–14/2010 wkw
 original article

Fig. 6. Levofloxacin-based triple therapy tolerance compared to bismuth-based quadruple therapy including the frequency of side effects of
treatment. Heterogeneity among studies

Fig. 7. Levofloxacin-based triple therapy tolerance compared to bismuth-based quadruple therapy including rates of discontinuation of therapy
due to adverse events. No heterogeneity among studies

Fig. 8. The eradication rates of levofloxacin-based triple therapy and bismuth-based quadruple therapy for patients with clarithromycin-resistant
H. pylori strains. No heterogeneity among studies

wkw 13–14/2010 © Springer-Verlag Second-line treatment for Helicobacter pylori infection 419
 original article
Discussion
In the present study, when clarithromycin-based triple
therapy was compared with bismuth-based quadruple
therapy, we have shown that clarithromycin-based quad-
ruple therapy appeared inferior to the latter (OR = 0.53,
95% CI: 0.35–0.80, P = 0.002). Antimicrobial resistance is
largely responsible for the poor eradication rates with clari-
thromycin-based triple therapy [38, 40]. Clarithromycin-
resistant strains of H. pylori are prevalent around the
world, with the average resistances rates as high as 24% in
Europe and 10% in the United States [38]. Clarithromycin,
a key component of many treatment regimens used to
eradicate H. pylori, is particularly sensitive to degradation
in an acidic environment and has a half-life of less than
1 hour in the stomach when the pH is 2 or lower [39]. The
gastric mucus layer serves as a mechanical barrier that
limits the distribution of the antimicrobials, and gastric
emptying limits retention of the antibiotics in the stomach
[39]. In addition, genetic mutations of H. pylori strains may
be involved in mechanisms of drug resistance [41, 42].
Many articles have reported that three point mutations
(A2143G, A2142G, and A2142C) were involved in H. pylori
clarithromycin resistance. In particular, A2143G and
A2142G transitions were the most prevalent point muta-
tions in Europe and the United States [43–44], while the
A2144G mutation was more frequent in Asia [45, 46]. More-
over, these different mutations affected H. pylori treatment
outcome [41]. The current background resistance rates of
H. pylori to metronidazole in many countries are also high.
Almost all studies indicate that resistances to both metro-
nidazole and clarithromycin are important factors leading
to treatment failure. Metronidazole resistance plays a par-
ticularly important role in the significant failure rate of
metronidazole-containing salvage therapy. But, when the
eradication rates for patients with metronidazole-resistant
were compared, there was no significant difference
between levofloxacin-based triple therapy and bismuth-
based (mainly metronidazole-containing therapy) quad-
ruple therapy. It has been reported that quadruple therapy
consisting of a similar combination of antibiotics and a
proton-pump inhibitor has repeatedly been shown to be
equally effective in spite of the presence of metronidazole
resistance when the attention was given to the dose of
metronidazole (e.g. 400 or 500 mg three times daily) and
duration of therapy of approximately 14 days [47–52]. So,
metronidazole-containing quadruple therapy may have
overcome metronidazole resistance to some degree in our
studies. Overall, because post-treatment antibiotic resist-
ance to metronidazole and clarithromycin is common, it
has been suggested that therapy should not combine clari-
thromycin and metronidazole [53].
Second-generation fluoroquinolones (moxifloxacin
and levofloxacin) have proved effective in different proto-
cols. Moxifloxacin and levofloxacin are fluoroquinolone
antibacterial agents with a broad spectrum of activity
against Gram-positive and Gram-negative bacteria and
proven efficacy in the treatment of infections of the respi-
ratory tract, genitourinary tract, digestive tract, and skin
[31, 54]. They have many advantages in terms of safety and
420 Second-line treatment for Helicobacter pylori infection
simplicity dosing management [33, 55]. Although our me-
ta-analysis suggested that the eradication rates of the
moxifloxacin-based triple therapy has a slight superiority
to bismuth-based quadruple therapy (OR = 1.78, 95% CI:
0.98–3.22, P = 0.06), there was no significant difference be-
tween them. However, more studies are needed to cofirm
it. When we separately compared 7-day levofloxacin-
based triple therapy with 7-day bismuth-based regimen,
the eradication rates also demonstrated no superiority of
7-day levofloxacin-based triple therapy over the latter
(OR = 1.04, 95% CI: 0.78–1.39, P = 0.81). But the 10-day levo-
floxacin-based triple therapy has the superiority over
7-day bismuth-based quadruple therapy (OR = 4.79, 95%
CI: 2.95–7.79, P < 0.00001). In addition, the levofloxacin-
based triple therapy was simple and well tolerated in the
present study. However, resistance to quinolones is easily
acquired. The use of moxifloxacin and levofloxacin should
be confined to ‘rescue’ therapy, in order to avoid rapidly
increasing H. pylori resistance toward such antibiotics
[56]. Nevertheless, some studies reported that levofloxacin
resistance to H. pylori could not be overcome by the use of
high-dose levofloxacin (1000 mg daily) regimen [23]. The
further studies need to be certified.
Furthermore, when we compared quadruple therapy
(7 vs. 14 days), we found that there was no significant effect
from increasing the duration of treatment. Compliance is
one of the important determinants of the outcome of treat-
ment [14]. It has been reported that compliance in 14-day
durations of treatment is usually worse than 7 days. Also,
7 days duration of treatment was associated with fewer
side effects [29], and produced equally good results com-
pared to 14 days duration of treatment. However, we need
large sample to confirm the conclusion.
In conclusion, second-generation fluoroquinolone-
based triple therapy, especially 10-day levofloxacin-based
triple therapy, can be suggested as the regimen of choice
for rescue therapy in the eradication of persistent H. pylori
infection.
Acknowledgment
This work was supported by Natural Science Funds of
China (No. 30770992), and Social Development Funds of
Jiangsu Province, China (No. B52007070 and H200702).
Conflict of interest
The authors declare no conflict of interest.
References
1. Buckley M, O’Morain C (1995) Prevalence of Helicobacter
pylori in non-ulcer dyspepsia. Aliment Pharmacol Ther
9[Suppl 2]: 53–8 [PMID:8547529]
2. Marshall BJ, Warren JR (1984) Unidentified curved bacilli in
the stomach of patients with gastritis and peptic ulceration.
Lancet 1: 1311–5 [PMID:6145023]
3. Marshall BJ, Windsor HM (2005) The relation of Helicobacter
pylori to gastric adenocarcinoma and lymphoma: patho-
physiology epidemiology, screening clinical presentation
treatment, and prevention. Med Clin North Am 89: 313–44
[PMID:15656929]
© Springer-Verlag 13–14/2010 wkw

4. Graham DY, Lu H, Yamaoka Y (2007) A report card to grade
Helicobacter pylori therapy [Editorial]. Helicobacter 12:
275–8 [PMID:17669098]
5. Cristina L, Palazzuoli A, Giordano N (2006) H. pylori infec-
tion and systemic antibodies to CagA and heat shock protein
60 in patients with coronary heart disease. World J Gastroen-
terol 12(48): 7815–20
6. Fujimura K, Kuwana M, Kurata Y, Imamura M, Harada H,
Sakamaki H, Koda K, Nomura S, Sugihara S, Shimomura T,
Fujimoto TT, Oyashiki K, Ikeda Y (2005) Is eradication thera-
py useful as the first line of treatment in Helicobacter pylori-
positive idiopathic thrombocytopenic purpura? Analysis of
207 eradicated chronic ITP cases in Japan. Int J Hematol
81(2): 162–8 [PMID:15765787]
7. Grande M, Cadeddu F, Villa M, Attinà GM, Muzi MG, Nigro
C, Rulli F, Farinon AM (2008) Helicobacter pylori and gas-
troesophageal reflux disease. World J Surg Oncol 6: 74
[PMID:18601740]
8. Graham DY, Lew GM, Klein PD, Evans DG, Evans DJ Jr, Saeed
ZA, et al (1992) Effect of treatment of Helicobacter pylori in-
fection on the long-term recurrence of gastric or duodenal
ulcer. A randomized, controlled study. Ann Intern Med 116:
705–8 [PMID:1558340]
9. Malfertheiner P, Megraud F, O’Morain C, Bazzoli F, El-Omar
E, Graham D, Hunt R, Rokkas T, Vakil N, Kuipers EJ (2007)
Current concepts in the management of Helicobacter pylori
infection: the Maastricht III Consensus Report. Gut 56(6):
772–81 [PMID:17170018]
10. Gisbert JP (2008) “Rescue” regimens after Helicobacter py-
lori treatment failure. World J Gastroenterol 14(35): 5385–
402 [PMID:18803350]
11. Mégraud F (2004) H. pylori antibiotic resistance: preva-
lence, importance, and advances in testing. Gut 53: 1374–84
[PMID:15306603]
12. Broutet N, Tchamgoué S, Pereira E, Lamouliatte H, Salamon
R, Mégraud F (2003) Risk factors for failure of Helico-
bacter pylori therapy – results of an individual data analy-
sis of 2751 patients. Aliment Pharmacol Ther 17: 99–109
[PMID:12492738]
13. Malfertheiner P, Megraud F, O’Morain C, et al (2002) Current
concepts in the management of Helicobacter pylori infec-
tion – The Maastricht 2 – 2000 Consensus Report. Aliment
Pharmacol Ther 16: 167–80 [PMID:11860399]
14. Saad RJ, Schoenfeld P, Kim HM, Chey WD (2006)
Levofloxacin-based triple therapy versus Bismuth-based
quadruple therapy for persistent Helicobacter pylori infec-
tion: a meta-analysis. Am J Gastroenterol 101(3): 488–96
[PMID:16542284]
15. Chey WD, Wong BC (2007) Practice Parameters Committee
of the American College of Gastroenterology. American Col-
lege of Gastroenterology guideline on the management of
Helicobacter pylori infection. Am J Gastroenterol 102: 1808–
25 [PMID:17608775]
16. Gisbert JP, Morena F (2006) Systematic review and meta-
analysis: levofloxacin-based rescue regimens after Heli-
cobacter pylori treatment failure. Aliment Pharmacol Ther
23(1): 35–44 [PMID:16393278]
17. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ,
Gavaghan DJ, McQuay HJ (1996) Assessing the quality of
reports of randomized clinical trials: Is blinding necessary?
Control Clin Trials 17: 1–12 [PMID:8721797]
18. Kuo CH, Hu HM, Kuo FC, Hsu PI, Chen A, Yu FJ, Tsai PY, Wu
IC, Wang SW, Li CJ, Weng BC, Chang LL, Jan CM, Wang WM,
Wu DC (2009) Efficacy of levofloxacin-based rescue therapy
for Helicobacter pylori infection after standard triple thera-
py: a randomized controlled trial. J Antimicrob Chemother
63(5): 1017–24 [PMID:19246508]
19. Jung HS, Shim KN, Baik SJ, Na YJ, Kang MJ, Jung JM, Ha CY,
Jung SA, Yoo K (2008) Efficacy of levofloxacin-based tri-
ple therapy as second-line Helicobacter pylori eradication.
Korean J Gastroenterol 51(5): 285–90 [PMID:18516012] (ab-
stracts in English, full articles in Korean)
wkw 13–14/2010 © Springer-Verlag
original article
20. Zhang H, Yu J, Jia CH (2008) Rescue regimen after Helico-
bacter pylori eradication failure with triple therapy. Chin J
Clin Pharmacol 24(4): 361–3
21. Liang ZG, Quan HB, He SZ (2007) Clinical research on levo-
floxacin-based therapy curing Helicobacter pylori as a sal-
vage treatment. Hai Nan Yi Xue 18: 1–3
22. Zhang Y, Wu J, Jia YG (2007) Observation on effect of levo-
floxacin-based therapy as a salvage therapy to Helicobacter
pylori. Chin J Dig 19: 320–1
23. Wong WM, Gu Q, Chu KM, Yee YK, Fung FM, Tong TS,
Chan AO, Lai KC, Chan CK, Wong BC (2006) Lansoprazole,
levofloxacin and amoxicillin triple therapy vs. quadruple
therapy as second-line treatment of resistant Helicobacter
pylori infection. Aliment Pharmacol Ther 23(3): 421–7
[PMID:16423001]
24. Nista EC, Candelli M, Santoro M, et al (2005) Levofloxacin-
base triple therapy in second-line treatment for H. pylori
eradication: Update. Gastroenterology 128: A-63
25. Wong WM, Chu KM, Yee YK, et al (2004) Randomized con-
trolled trial of lansoprazole, levofloxacin and amoxycillin
triple therapy versus lansoprazole-based quadruple thera-
py as a rescue treatment after failure of Helicobacter pylori
(H. pylori) eradication with standard triple therapies. Gut
53[Suppl VI]: A-124
26. Nista EC, Candelli M, Fini L, et al (2004) 10 days levofloxacin-
based triple therapy in second-line treatment for Helico-
bacter pylori eradication. Gastroenterology 126[4 Suppl 2]:
A-74
27. Bilardi C, Dulbecco P, Zentilin P, Reglioni S, Iiritano E, Parodi
A, Accornero L, Savarino E, Mansi C, Mamone M, Vigneri S,
Savarino V (2004) A 10-day levofloxacin-based therapy in
patients with resistant Helicobacter pylori infection: a con-
trolled trial. Clin Gastroenterol Hepatol 2(11): 997–1002
[PMID:15551252]
28. Wong WM, Gu Q, Lam SK, Fung FM, Lai KC, Hu WH, Yee YK,
Chan CK, Xia HH, Yuen MF, Wong BC (2003) Randomized
controlled study of rabeprazole, levofloxacin and rifabutin
triple therapy vs. quadruple therapy as second-line treat-
ment for Helicobacter pylori infection. Aliment Pharmacol
Ther 17(4): 553–60 [PMID:12622764]
29. Perri F, Festa V, Merla A, Barberani F, Pilotto A, Andriulli A
(2003) Randomized study of different ‘second-line’ therapies
for Helicobacter pylori infection after failure of the standard
‘Maastricht triple therapy’. Aliment Pharmacol Ther 18(8):
815–20 [PMID:14535875]
30. Nista EC, Candelli M, Cremonini F, Cazzato IA, Di Caro S,
Gabrielli M, Santarelli L, Zocco MA, Ojetti V, Carloni E,
Cammarota G, Gasbarrini G, Gasbarrini A (2003) Levo-
floxacin-based triple therapy vs quadruple therapy in sec-
ond-line Helicobacter pylori treatment: a randomized trial.
Aliment Pharmacol Ther 18(6): 627–33 [PMID:12969089]
31. Kang JM, Kim N, Lee DH, Park YS, Kim YR, Kim JS, Jung
HC, Song IS (2007) Second-line treatment for Helicobacter
pylori infection: 10-day moxifloxacin-based triple therapy
versus 2-week quadruple therapy. Helicobacter 12(6): 623–8
[PMID:18001404]
32. Cheon JH, Kim N, Lee DH, Kim JM, Kim JS, Jung HC, Song
IS (2006) Efficacy of moxifloxacin-based triple therapy as
second-line treatment for Helicobacter pylori infection.
Helicobacter 11(1): 46–51 [PMID:16423089]
33. Bago J, Pevec B, Tomić M, Marusić M, Bakula V, Bago P
(2009) Second-line treatment for Helicobacter pylori infec-
tion based on moxifloxacin triple therapy: a randomized
controlled trial. Wien Klin Wochenschr 121(1–2): 47–52
[PMID:19263014]
34. Peitz U, Sulliga M, Wolle K, Leodolter A, Von Arnim U, Kahl
S, Stolte M, Börsch G, Labenz J, Malfertheiner P (2002) High
rate of post-therapeutic resistance after failure of mac-
rolide-nitroimidazole triple therapy to cure Helicobacter
pylori infection: impact of two second-line therapies in a
randomized study. Aliment Pharmacol Ther 16(2): 315–24
[PMID:11860415]
Second-line treatment for Helicobacter pylori infection 421
 original article
35. Qasim A, Sebastian S, Thornton O, Dobson M, McLoughlin
R, Buckley M, O’Connor H, O’Morain C (2005) Rifabutin-
and furazolidone-based Helicobacter pylori eradication
therapies after failure of standard first- and second-line
eradication attempts in dyspepsia patients. Aliment Phar-
macol Ther 21: 91–6 [PMID:15644050]
36. Peitz U, Sulliga M, Nusch A, Wolle K, Leodolter A, Stolte M,
Borsch G, Labenz J, Malfertheiner P (1998) Randomised
study on the impact of resistance on second-line treatment
after failed metronidazole-clarithromycin containing triple
therapy. Gut 43(2S) [Suppl]: 82A–3A
37. Mantzaris GJ, Petraki C, Petraki K, Roussos A, Karagiannidis
A, Christidou A, Christoforidis P, Polyzou P, Koilakou S,
Triantafyllou G (2005) Prospective, randomized study of
seven versus fourteen days omeprazole quadruple therapy
for eradication of Helicobacter pylori infection in patients
with duodenal ulcer after failure of omeprazole triple ther-
apy. Ann Gastroenterol 18(3): 330–5
38. Vakil N, Vaira D (2008) Sequential therapy for Helicobacter
pylori: time to consider making the switch? JAMA 300:
1346–7 [PMID:18799448]
39. Vakil N, Megraud F (2007) Eradication therapy for Helicobacter
pylori. Gastroenterology 133(3): 985–1001 [PMID:17854602]
40. Marshall B, MD (2008) Sequential therapy for Helicobacter
pylori: a worthwhile effort for your patients. Ann Intern Med
148: 962–3 [PMID:18490666]
41. De Francesco V, Margiotta M, Zullo A (2006) Clarithromy-
cin-resistant genotypes and eradication of Helicobacter py-
lori. Ann Intern Med 144: 94–100 [PMID:16418408]
42. Owen RJ (2002) Molecular testing for antibiotic resistance in
Helicobacter pylori. Gut 50: 285–9 [PMID:11839700]
43. van Doorn LJ, Glupczynski Y, Kusters JG, Mégraud F, Midolo
P, Maggi Solcà N, et al (2001) Accurate prediction of mac-
rolide resistance in Helicobacter pylori by a PCR line probe
assay for detection of mutations in the 23S rRNA gene: mul-
ticenter validation study. Antimicrob Agents Chemother 45:
1500–4 [PMID:11302817]
44. Versalovic J, Osato MS, Spakovsky K, Dore MP, Reddy R,
Stone GG, et al (1997) Point mutations in the 23S rRNA gene
of Helicobacter pylori associated with different levels of clar-
ithromycin resistance. J Antimicrob Chemother 40: 283–6
[PMID:9301997]
45. Kaneko F, Suzuki H, Hasegawa N, Kurabayshi K, Saito H,
Otani S, et al (2004) High prevalence rate of Helicobacter py-
422 Second-line treatment for Helicobacter pylori infection
lori resistance to clarithromycin during long-term multiple
antibiotic therapy for chronic respiratory disease caused by
non-tuberculous mycobacteria. Aliment Pharmacol Ther
20[Suppl 1]: 62–7 [PMID:15298607]
46. Masaoka T, SuzukiKurabayashi HK, Kamiya AG, Ishii
H (2004) Second-line treatment of Helicobacter pylori
infection after dilution agar methods and PCR-RFLP
analysis. Aliment Pharmacol Ther 20[Suppl 1]: 68–73
[PMID:15298608]
47. de Boer W, Driessen W, Jansz A, Tytgat G (1995) Effect of
acid suppression on efficacy of treatment for Helicobacter
pylori infection. Lancet 345: 817–20
48. de Boer W, Driessen WM, Potters VP, Tytgat GN (1994) Ran-
domized study comparing 1 with 2 weeks of quadruple ther-
apy for eradicating Helicobacter pylori. Am J Gastroenterol
89: 1993–7
49. de Boer W, Tytgat GN (1995) 90% cure: which anti-Helico-
bacter therapy can achieve this treatment goal? Am J Gastro-
enterol 90: 1381–2
50. de Boer W, van Etten RJ, Lai JY, Schneeberger PM, van de
Wouw BA, Driessen WM (1996) Effectiveness of quadruple
therapy using lansoprazole, instead of omeprazole, in curing
Helicobacter pylori infection. Helicobacter 1: 145–50
51. de Boer W (1996) How to achieve a near 100% cure rate for
H. pylori infection in peptic ulcer patients. A personal view-
point. J Clin Gastroenterol 22: 313–6
52. de Boer W, van Etten RJ, van de Wouw BA, Schneeberger
PM, Van Oijen AH, Jansen JB (2000) Bismuth-based quad-
ruple therapy for Helicobacter pylori – a single triple capsule
plus lansoprazole. Aliment Pharmacol Ther 14: 85–9
53. Gisbert JP, Pajares JM (2002) Helicobacter pylori rescue regi-
men when proton pump inhibitor-based triple therapies fail.
Aliment Pharmacol Ther 16(6): 1047–57 [PMID:12030945]
54. Croom KF, Goa KL (2003) Levofloxacin: a review of its use
in the treatment of bacterial infections in the United States.
Drugs 63: 2769–802 [PMID:14664657]
55. Nista EC, Candelli M, Zocco MA, Cremonini F, Ojetti V,
Finizio R, Spada C, Cammarota G, Gasbarrini G, Gasbarrini
A (2006) Levofloxacin-based triple therapy in first-line treat-
ment for Helicobacter pylori eradication. Am J Gastroenterol
101: 1985–90 [PMID:16968503]
56. Gisbert JP, Castro-Fernandez M, Bermejo F, et al (2006)
Third-line rescue therapy with levofloxacin after two H. py-
lori treatment failures. Am J Gastroenterol 101: 243–7
© Springer-Verlag 13–14/2010 wkw