Article Text

Download PDFPDF

Original article
Ten day sequential versus 10 day modified bismuth quadruple therapy as empirical firstline and secondline treatment for Helicobacter pylori in Chinese patients: an open label, randomised, crossover trial
  1. Kevin S H Liu,
  2. Ivan F N Hung,
  3. W K Walter Seto,
  4. Teresa Tong,
  5. Axel S J Hsu,
  6. Frank Y F Lam,
  7. David Y K But,
  8. S Y Wong,
  9. Wai K Leung
  1. Department of Medicine, Queen Mary Hospital, University of Hong Kong, Hong Kong, Hong Kong
  1. Correspondence to Dr W K Leung, Department of Medicine, 4/F, Professorial Block, Queen Mary Hospital, Pokfulam Road, Hong Kong, Hong Kong; waikleung{at}


Objective Treatments with sequential therapy (SEQ) or bismuth quadruple (QUAD) therapy have been proposed as empirical firstline regimens for Helicobacter pylori. We compared the efficacy and tolerability of 10 day SEQ with 10 day modified QUAD as both firstline and secondline treatments for H pylori in a randomised crossover study.

Design H pylori positive and treatment naïve patients were randomly assigned to receive either 10 day SEQ (esomeprazole for 10 days, amoxicillin for an initial 5 days, followed by clarithromycin and metronidazole for a subsequent 5 days) or modified QUAD (esomeprazole, bismuth subcitrate, tetracycline and metronidazole). H pylori eradication was confirmed by urea breath test at 8 weeks. Patients who failed the initial assigned treatment were crossed over to receive the alternate regimen. The primary outcome was eradication rates of firstline treatment by intention to treat (ITT) and per protocol (PP) analyses.

Results 357 patients were randomised to receive either SEQ or QUAD. The PP eradication rates of the SEQ and QUAD groups were 95.2% and 98.8%, respectively (p=0.10). Based on ITT analysis, the corresponding eradication rates were 89.4% and 92.7%, respectively (p=0.36). Eight (4.8%) patients in the SEQ and two (1.2%) patients in the QUAD who failed the firstline treatment were crossed over to the alternate regimen with 100% retreatment success. The overall incidence of adverse events was higher in the QUAD (16.7%) than in the SEQ (8.1%; p=0.032) group.

Conclusions Ten day sequential and modified bismuth quadruple therapies are both highly effective as empirical firstline therapies for H pylori in Chinese patients. NCT 01760824

  • Helicobacter Pylori - Treatment
  • 13C-Urea Breath Test
  • Antibiotics
View Full Text

Statistics from

Significance of this study

What is already known on this subject?

  • The success rate of the current clarithromycin containing triple therapy has declined substantially due to increasing antimicrobial resistance.

  • Alternative regimens, including bismuth containing quadruple therapy or non-bismuth containing quadruple therapy (either sequential or concomitant), have been recommended as firstline therapies by the latest Maastricht Consensus, particularly in areas with a high prevalence of clarithromycin resistance.

  • There is no direct comparison of these two alternative regimens as firstline treatments for Helicobacter pylori.

What are the new findings?

  • Both 10 day modified bismuth quadruple and 10 day sequential therapies were found to have a very high eradication rate (>95% by per protocol analysis) when used a firstline treatment for H pylori in Hong Kong Chinese patients.

  • Patients who failed the initial assigned treatment (either sequential or bismuth quadruple) were all successfully eradicated when crossed over to the alternative regimen empirically without the need for antimicrobial testing.

How might it impact on clinical practice in the foreseeable future?

  • Both 10 day modified bismuth quadruple and 10 day sequential therapy can be recommended as empirical firstline treatment for H pylori in Hong Kong.

  • Crossover treatment between modified bismuth quadruple and sequential therapy appears to be an effective empirical secondline therapy when firstline therapy fails.


The identification of Helicobacter pylori by Warren and Marshall1 in 1983 has revolutionised our understanding of gastroduodenal diseases. It has become clear that chronic infection with H pylori not only leads to chronic gastritis but also peptic ulcer diseases and gastric malignancy, including gastric adenocarcinoma and gastric mucosa associated lymphoid tissue lymphoma. H pylori was therefore classified by the International Agency for Research on Cancer of the WHO as a class 1 carcinogen.2 Eradication of H pylori leads to long term cure of peptic ulcer, possibly a cure for mucosa associated lymphoid tissue lymphoma and may even reduce the risk of gastric cancer development if given early.3

Clarithromycin containing triple therapy consists of a proton pump inhibitor, clarithromycin and amoxicillin, has been the standard firstline eradication regimen since its introduction in 1996.4 It is still recommended by the Maastricht IV consensus as the firstline treatment in regions with low prevalence of clarithromycin resistance.5–7 However, the eradication rate of clarithromycin containing triple therapy is declining globally to an unacceptable eradication rate of less than 80%,8 and even less than 60% in some countries,9 ,10 largely due to the rising clarithromycin resistance.11 There are wide geographic variations in clarithromycin resistance rates in different countries, ranging from 5.2% in Belgium to 55.6% in Japan.12 In Hong Kong, the resistance rate to clarithromycin increased from 4.7% in 1997 to 8.3% in 200113 and to 13% in 2005.14

In response to the falling H pylori eradication rate due to rising clarithromycin resistance, several regimens, including sequential therapy, concomitant therapy, hybrid therapy and levofloxacin containing therapy, have been studied with variable success.15–18 The latest Maastricht IV Consensus Report recommended using bismuth containing quadruple therapy or sequential therapy as an alternative empirical firstline treatment, particularly in areas with high clarithromycin resistance.7 However, there are no direct comparisons between bismuth containing quadruple therapy and sequential therapy as firstline treatment for H pylori.

The aim of this study was to compare the eradication success rates of 10 day sequential therapy (SEQ) with a modified 10 day bismuth containing quadruple therapy (QUAD) as an empirical firstline treatment for H pylori in Hong Kong Chinese patients with a background of rising clarithromycin resistance. We further determined the role of these two regimens as empirical secondline therapy in patients who failed the initial assigned regimen in a crossover design.

Patients and methods

Study design and patients

This was a prospective, open label, randomised, crossover study performed in the Department of Medicine, Queen Mary Hospital, Hong Kong. Eligible patients were recruited from patients who were referred to our department for evaluation of dyspeptic symptoms by upper gastrointestinal endoscopy between July 2011 and June 2012. All patients would have undergone upper gastrointestinal endoscopy prior to recruitment into the study. Patients were eligible if they were aged 18 years or older and had confirmed H pylori infection. H pylori infection was diagnosed by either a positive rapid urease test or positive identification of H pylori in a histological sample using haematoxylin and eosin and/or Giemsa stain. Patients were excluded if they had any of the followings: previous eradication therapy for H pylori; known allergies to any of the study medications; recent use of antibiotics, bismuth salts, probiotic preparations, proton pump inhibitors or non-steroidal anti-inflammatory drugs in the preceding 1 month; pregnant or lactating women; previous gastric surgery; alcohol or drug abuse; or severe concurrent medical illnesses, such as liver failure, renal failure or terminal malignancy.

All patients provided written informed consent for participation in the study. The study protocol was approved by the institutional review board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster.

Randomisation and treatment allocation

Eligible patients were randomised in a 1:1 ratio to receive either 10 day SEQ or 10 day QUAD. Sequential therapy consisted of esomeprazole 20 mg twice daily and amoxicillin 1 g twice daily for the first 5 days, followed by esomeprazole 20 mg twice daily, clarithromycin 500 mg twice daily and metronidazole 400 mg four times daily for the subsequent 5 days. Quadruple therapy comprised esomeprazole 20 mg twice daily, bismuth subcitrate 120 mg four times daily, tetracycline 500 mg four times daily and metronidazole 400 mg four times daily for a total of 10 days. We used computer generated random sequences in blocks of four. All randomisation codes were put into sealed opaque envelopes and kept by an independent research assistant. After obtaining informed consent, the investigators would call the independent research assistant to open the envelope for the allocated regimen for that patient.

Written instructions were provided to all patients on how to take the study medications and their common side effects to reinforce drug compliance. All adverse events were recorded in a specific questionnaire by the research personnel. Drug compliance was evaluated by counting the number of unused study medications, and non-compliance was defined as failure to take 90% of the study medications.

Follow-up urea breath test

The success of H pylori eradication was determined by the 13C-urea breath test (UBT) at 8 weeks after completion of the assigned treatment. 13C-UBT was performed as described previously.19 Briefly, patients were fasted for at least 6 h before obtaining the baseline breath sample. The patient then swallowed one diabact UBT tablet (50 mg of 13C-urea and 456 mg of anhydrous citric acid; Orexo AB, Uppsala, Sweden) with 200 mL of water. A second breath sample was obtained 10 min later. All of the breath samples were analysed by a purpose built isotope ratio mass spectrometer (FANCi2, the Fischer ANalysen Instrumente GmbH, Leipzig, Germany) by a technician who was blinded to treatment allocation. The cut-off value used was 2.5%.19

Crossover for treatment failure

Patients who failed the initial assigned eradication regimen, as defined by a persistent positive UBT at the end of 8 weeks, were crossed over to receive the alternate treatment regimen. Patients who failed SEQ were given QUAD, and those who failed QUAD were given SEQ as secondline therapy. A second UBT were performed at 8 weeks after completion of the second regimen.


The primary outcome of this study was the H pylori eradication rates of the initial assigned treatment by intention to treat (ITT) and per protocol (PP) analyses. The secondary outcomes were H pylori eradication rates in the crossover treatment, the frequency of adverse events and treatment compliance.

Sample size and statistical analysis

We estimated the sample size of this study based on two recent meta-analyses on H pylori eradication studies.15 ,20 We assumed the eradication rate of the SEQ group to be superior to the QUAD group, and the eradication rates of the SEQ group and QUAD group to be 90% and 78%, respectively. With a power of 80% and a two sided type 1 error of 5%, 150 patients were needed for each treatment arm that would also allow for 10% loss to follow-up.

Efficacy of H pylori eradication was compared by both ITT and PP analyses. All randomised patients were included in the ITT analysis. Patients who did not return for UBT were treated as treatment failures in the ITT analysis. Patients who violated the study protocol or were non-compliant with the assigned treatment were excluded from the PP analysis. A modified ITT analysis was performed to include all patients who returned for follow-up UBT, irrespective of whether the assigned treatment was completed. Continuous variables were expressed as mean (SD) and compared by Student's t test. Categorical data comparisons were performed by Pearson's χ2 test or Fisher's exact test where appropriate. A two sided p value of <0.05 was considered statistically significant. All statistical analysis was performed by SPSS V.20.0 software for Windows (SPSS Inc, Chicago, Illinois, USA).



A total of 423 patients were screened within the 12 month study period and 66 were not recruited for various reasons, as listed in figure 1. A total of 357 H pylori infected patients were recruited: 179 patients were assigned to the SEQ group and 178 patients were assigned to the QUAD group. All patients had undergone upper gastrointestinal endoscopy with H pylori infection confirmed. Baseline characteristics and endoscopy findings of the two groups are shown in table 1. There were more women (56.6%) than men in this study, and mean age of the patients was 56.4 years. There was no significant difference in baseline demographics or endoscopy findings between the two groups. Thirteen (7.3%) patients in the SEQ group and 15 (8.4%) patients in the QUAD group did not complete the study. The details are listed in figure 1.

Table 1

Baseline characteristics of the study patients

Figure 1

Trial profile. ITT, intention to treat; PP, per protocol; QUAD, modified quadruple therapy; SEQ, sequential therapy; UBT, urea breath test.

Eradication rates of firstline treatment

The overall eradication rates by ITT and PP analysis are summarised in table 2. Based on ITT analysis, eradication rates in the SEQ group were 89.4% (95% CI 84% to 93.1%), and 92.7% (95% CI 87.9% to 95.7%) in the QUAD group (difference=−3.3%, 95% CI −9.5% to 2.8%; p=0.36). By PP analysis, the respective eradication rates in the SEQ and QUAD groups were 95.2% (95% CI 90.8% to 97.5%) and 98.8% (95% CI 95.6% to 99.7%); the difference was −3.6% (95% CI −8.0% to 0.3%; p=0.10). We further performed a modified ITT analysis, which included all patients who returned for follow-up UBT. Based on this modified ITT analysis, the eradication rates in the SEQ and QUAD groups were 94.7% (95% CI 90.2% to 91.7%) and 98.8% (95% CI 95.7% to 99.7%); The difference was −4.1% (95% CI −9.1% to 0.03%; p=0.06).

Table 2

Helicobacter pylori eradication rates in the sequential therapy and quadruple therapy groups

Subgroup analysis was also performed to determine the eradication rates in patients with and without gastroduodenal ulcers. The results are listed in table 3. There was no significant difference in eradication rates between the QUAD and SEQ groups in patients with or without gastroduodenal ulcers.

Table 3

Helicobacter pylori eradication rates of the sequential treatment and modified quadruple treatment groups according to endoscopic findings

Crossover secondline treatment

Ten patients who failed H pylori eradication after completing the initial assigned treatment regimen were available for crossover treatment by the alternate regimen. Eight patients in the SEQ group were given secondline treatment with QUAD and two patients in the QUAD group were offered SEQ. All 10 patients had a negative UBT result 8 weeks after the second course of treatment.

When combining the analysis of both firstline and secondline treatments, the PP eradication rate of H pylori by SEQ was 95.2% (95% CI 90.5% to 97.8%) and by QUAD was 98.8% (95% CI 95.4% to 99.8%). The difference was −3.6% (95% CI −8.4% to 0.7%; p=0.06), which was not statistically significant.

Compliance and adverse events

The compliance rate of the study patients was 92.7% and 91.6% in the SEQ and QUAD groups, respectively (difference=1.1%, 95% CI −0.55% to 1.96%; p=0.68). The frequencies of adverse events in this study are listed in table 4. Overall, there were significantly more adverse events in the QUAD group (16.7%) than in the SEQ group (8.1%; p=0.032). The most common adverse events reported in the QUAD group were nausea (3.4%), vomiting (2.3%) and skin rash (2.3%). In the SEQ group, skin rash (2.3%) and taste disturbance (1.7%) were the most commonly reported adverse effects. All reported side effects were mild and resolved spontaneously after stopping treatment.

Table 4

Adverse events of the study patients

We further determined the potential risk factors associated with failed eradication by binary logistic regression. However, none of the factors, including patient age, gender, alcohol and smoking status, presence of gastrodudoenal ulcer, presence of adverse effects and treatment assignment, was associated with eradication failure.


Both sequential therapy and bismuth containing quadruple therapy are recommended as the alternative firstline treatment for H pylori in the Maastricht IV Consensus, particularly in regions with a high prevalence of clarithromycin resistance.7 This is, to our knowledge, the first prospective, randomised, crossover trial that directly compared 10 day sequential therapy with 10 day modified bismuth containing quadruple therapy as both firstline and secondline treatments for H pylori. Our results showed that both regimens were highly effective in eradicating H pylori, with a more than 95% eradication rate according to PP analysis. Even by ITT analysis, H pylori eradication rates of both regimens were more than 89%. In particular, bismuth quadruple therapy had a PP eradication rate of 98.8%. The overall eradication rate of bismuth quadruple therapy was about 3% higher than sequential therapy but the difference did not reach statistical significance. These results are highly promising and can be regarded as grade A standard, as proposed recently.21

The longer duration and additional antimicrobial agents may account for the higher eradication rates of both SEQ and QUAD quadruple regimens compared with conventional clarithromycin containing regimens. The eradication rates of the modified bismuth quadruple therapy in this study appeared to be much higher than the recent meta-analysis that showed a pooled ITT eradication rate of bismuth based quadruple therapy to be 78.3%.20 However, this meta-analysis included nine randomised controlled studies with different drug dosages and treatment durations. In contrast, we used a higher dose of metronidazole, given at 400 mg four times per day, for a longer duration of 10 days. Despite the usual side effects of bismuth treatment, the compliance of our study patients was very high, with more than 90% of our patients completing the assigned treatment. Recent studies from China also suggested that the use of full dose bismuth quadruple therapies are highly effective in Chinese patients, even in the presence of prior treatment failures and antibiotics resistance.22 ,23

Although the eradication rate of the 10 day sequential therapy was slightly lower than the bismuth containing quadruple therapy, it was comparable with the results published by the recent meta-analysis, with an ITT eradication rate of 92.4%.15 The results of the sequential therapy in Asia were however variable, ranging from 81.9% in Korea,24 to 82.6% in China25 and to 90.5% in Taiwan.26 The recent study from Taiwan showed that extending sequential therapy from 10 to 14 days slightly increased the eradication rate by 3–4%.26

In this study, we also determined the role of sequential therapy or bismuth quadruple therapy as secondline treatment for patients who failed the assigned firstline treatment. With the high success rate of both regimens as firstline therapy, only 10 (2.8%) patients were available for this crossover study. All 10 patients had H pylori eradicated after the alternative regimen. With the minimal overlap of antibiotics of the two regimens, patients who fail the initially assigned treatment could be conveniently allocated to the alternative treatment without the need for antibiotics susceptibility testing, which are not widely available. The combination of sequential therapy and bismuth quadruple therapy, in either order, may prove to be a very useful approach to firstline and secondline H pylori therapy in our daily practices.

With the availability of new single (three in one) capsule containing bismuth subcitrate, metronidazole and tetracycline,27 the number of tablets needs to be taken in the bismuth quadruple group will be substantially decreased and this will also likely increase drug compliance. The use of concomitant rather than sequential therapy may also simplify the complicated dosing regimen. To support this, a recent Spanish study showed a marginal 5% increase in eradication rate of concomitant over sequential therapy.28

In this study, the QUAD group had a statistically significant higher incidence of adverse events compared with the SEQ group, which was not unexpected. The reported adverse events of both QUAD and SEQ therapies were however usually mild and self-limiting, and did not have a major impact on the compliance of this study.

Our study has several limitations. First, pretreatment antimicrobial resistance of H pylori was not tested. The antimicrobial resistance profiles of our patients were only based on recent local data, and the results would definitely be beneficial if data on antimicrobial resistance were also available. Second, we used only one positive test result for diagnosis of H pylori infection in this study. As false positive biopsy based H pylori testing is infrequent, this may have an impact on patient recruitment but should not interfere with the estimation of the success rate of the eradication regimens. Third, the present study was set up as a superiority trial based on the results of previous studies suggesting that the eradication rates of the SEQ regimen were higher than those of the QUAD regimen. This assumption and sample size estimation were based on two recently published meta-analyses,15 ,20 which may be subject to flaws. In particular, the eradication rates of bismuth quadruple therapy as reported20 might underestimate the eradication rate of bismuth quadruple therapy. An earlier meta-analysis29 reported a much higher eradication rate of quadruple therapy (more than 85%) if it contained omeprazole and was given for 10–14 days. We found that both regimens were highly effective, with a more than 95% eradication rate based on PP analysis, and QUAD therapy appeared to have a higher success rate than SEQ therapy. Given the excellent eradication rates of both treatment arms (>95% according to PP analysis), demonstration of a statistically significant difference would require a very large sample size and may not confer any additional impact on our daily practices. Fourth, due to the small number of patients who failed H pylori eradication, the current crossover design for assessment of the success rate of secondline therapy was under powered. With the very high success rate of the current two regimens, a multicentre study will be needed to compare these two regimens as secondline therapies. On the other hand, it demonstrates the possible role of alternating these two regimens empirically in patients who fail the assigned firstline eradication therapy without the need for further antimicrobial resistance testing. Lastly, this study recruited only Chinese patients. Due to the potential differences in CYP2C19 genotypes30 and antimicrobial resistance patterns among different ethnic groups, further studies in other populations may be necessary to confirm our findings.

In conclusion, both sequential therapy and modified bismuth containing quadruple therapy were highly effective as firstline therapies in eradicating H pylori in Hong Kong Chinese patients and with a low frequency of adverse events.


We would like to thank all of the endoscopists and endoscopy nurses in Queen Mary Hospital for patient recruitment.


View Abstract


  • Contributors KSHL and IFNH: study concept and design, acquisition of the data, statistical analysis, data interpretation and drafting of the manuscript. WKWS, ASJH, FYFL, DYKB and SYW: patient recruitment, acquisition of the data and critical revision of the manuscript. TT: technical support and study coordination. WKL: study supervision, data interpretation, and drafting and critical revision of the manuscript.

  • Funding This study was supported by funding from the Division of Gastroenterology and Hepatology, Department of Medicine, University of Hong Kong.

  • Ethics approval The study was approved by the institutional review board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.