Dr Z. H. Ran, Department of Gastroenterology, Ren ji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai Institute of Digestive Disease, Shanghai 200001, China. Tel.: +86 21 63260930; fax: +86 21 63266027; e-mail: email@example.com
Background: As standard triple therapies of achieve unsatisfactory eradication of Helicobacter pylori, several alternative regimens have been proposed.
Objectives: To systematically evaluate whether sequential treatment eradicates H. pylori infection better than standard triple therapies and compare the risk of adverse events with these two regimens.
Methods: We searched electronic databases up to February 2008 for studies evaluating the efficacy of the 10-day sequential therapy vs. standard triple regimens for eradication of H. pylori. The pooled risk ratios (RR) and 95% confidence intervals (95% CI) were calculated.
Results: We identified 11 randomized trials, including eight full-text manuscripts and three abstracts. Pooled analysis demonstrated clear superiority of the sequential therapy over 7-day triple regimen with an RR of 1·23 (95% CI 1·19–1·27), and over 10-day triple regimen with a RR of 1·16 (95% CI 1·10–1·23). Adverse event rates were similar. For sequential therapy vs. 7-day triple therapies, RR = 0·96, 95% CI 0·70–1·31.
Conclusions: Sequential therapy was associated with a higher eradication rate of H. pylori compared with both 7-day triple regimen and 10-day triple regimen.
Helicobacter pylori are a global human pathogen and are the major cause of gastritis and the gastritis-associated diseases: gastric ulcer, duodenal ulcer, gastric cancer and mucosa-associated lymphoid tissue lymphoma (1). The indications for H. pylori eradication are steadily increasing and the widespread adoption of a test-and-treat strategy for the management of dyspepsia necessitates an eradication regimen that is affordable, highly effective and with few side-effects. Triple therapies including proton pump inhibitor (PPI) or ranitidine bismuth citrate, clarithromycin, amoxicillin or nitroimidazolic compounds, are first-line regimens administered for either 7 or 14 days as recommended by European and American guidelines, respectively (2, 3). It has become clear that the eradication rates of 7- or 10-day standard triple therapies are now typically below 80% mainly because of increasing primary clarithromycin resistance (4). For this reason, several alternative regimens have been proposed. Recently, a novel 10-day sequential regimen has been reported to have a very high H. pylori eradication rate. The sequential regimen is a simple dual therapy with a PPI plus 1 g amoxicillin (both twice daily) given for the first 5 days followed by a triple therapy including a PPI, 500 mg clarithromycin and 500 mg tinidazole (all twice daily) for the remaining 5 days (5). Such a therapeutic regimen consistently achieves a very high eradication rate with few side-effects in children, adult and elderly patients (6–16). A pooled analysis by Zullo et al. (17) also demonstrated that the 10-day sequential treatment regimen achieves higher eradication rates than standard triple therapies. Therefore, sequential therapy is very promising but remains an experimental therapy. To determine whether sequential treatment eradicates H. pylori infection better than standard triple therapies, we pool data from a wider range of previously published clinical studies.
We searched the following electronic databases: MEDLINE (1966 to February 2008), EMBASE (1980 to February 2008), the Cochrane Central Register of Controlled Trials (Cochrane Library Issue 1, 2008) and the Chinese Biomedical Database (1981 to February 2008) using the terms ‘Helicobacter pylori’ as medical subject headings and ‘Helicobacter pylori’, ‘H. pylori’, ‘7 days’, ‘seven days’, ‘10 days’, ‘ten days’ and ‘sequential’ as text words. There was no restriction on language. In addition, we manually screened reference lists of identified articles, retrieved reports citing relevant articles through Science Citation Index (1981 to February, 2008). Further searches were carried out in two trials registers (http://www.clinicaltrials.gov/ and http://www.controlled-trials.com/). Abstracts of major gastroenterological meetings, such as the Digestive Disease Week of the American Gastroenterological Association, the World Congress of Gastroenterology and European Helicobacter Study Group were reviewed for the past 5 years. We also contacted authors and experts in the specialty. The results of the computerized search were independently reviewed by two reviewers (Tong and Shen). Disagreements were resolved by third party adjudication (Ran).
Inclusion criteria and exclusion criteria
Studies were included if they were randomized controlled trials (RCTs) that compared sequential treatment with 7- or 10-day standard triple therapies for H. pylori eradication.
Inclusion criteria were as follows:
(a) Patients with H. pylori infection; H. pylori status must have been determined by at least one of following methods: rapid urease test, Helicobacter pylori stool antigen test (HpSA), histology or urea breath test before treatment and at least 4 weeks after the end of treatment;
(b) Intention-to-treat analysis;
(c) Studies must include at least two branches of treatment consisting of sequential therapy and standard triple therapy. Sequential regimens should include a PPI and amoxicillin for the first 5 days followed by a triple therapy including a PPI, clarithromycin and tinidazole for the remaining 5 days; triple regimen should be combinations of a PPI plus two antibiotics.
(a) Previous H. pylori eradication treatment;
(b) Use of PPIs, H2-receptor antagonists, bismuth preparations and antibiotics in the previous 4 weeks;
Data were abstracted using a standardized form. Abstracted information included patient characteristics (mean age, proportion of women and smoking habit), year of publication, publication status, trial design, treatment regimens, eradication rates for corresponding regimens, number of patients experiencing any adverse events, number of patients who withdrew because of adverse events, inclusion and exclusion criteria for enrollment, diagnostic methods of testing H. pylori infection before enrolling and after the end of treatment, number of patient with peptic ulcer/non-ulcer dyspepsia or with clarithromycin/metronidazole resistance and time for confirmation of H. pylori eradication after completing regimen.
Trial quality was assessed using the Jadad score system based on three items: randomization, double blinding, and description of withdrawals and dropouts, to generate scores from 0 to 5 (18). We considered trials to be of low quality when their scores were lesser than 3. The selected studies were scored independently by two investigators (Tong and Shen), and if there were disagreements, a third reviewer (Ran) was consulted to each the final scores. Inter-observer agreement between reviewers in the rating of papers was calculated and reported using the kappa statistic (19).
The primary outcome measures were the risk ratios (RR) for successful H. pylori eradication comparing sequential treatment with 7- or 10-day standard triple therapies. A pooled analysis was also planned to access several secondary outcomes, including (i) adverse events during H. pylori eradication; (ii) eradication rate of H. pylori for peptic ulcer dyspepsia; (c) eradication of H. pylori for non-ulcer dyspepsia; (d) eradication rate of H. pylori in patients with antibiotics resistance. Eradication rates were considered both on an intention-to-treat and on a per protocol basis.
Risk ratio and 95% confidence intervals (95% CI) were used as summary statistics. The pooled RR was calculated first by using a fixed-effect model with the Mantel–Haenszel method. The Breslow–Day test was used to examine statistical heterogeneity across the studies (P < 0·1) and I2 analysis was used to estimate the magnitude of variation attributable to heterogeneity rather than to chance. The DerSimonian and Laird random effect model was additionally applied to calculate pooled RR in case of significant heterogeneity across studies. In trials with three or more arms, we halved both the number of events and the total number of patients, e.g. comparison between sequential therapy and 7- or 10-day standard regimens for eradication of H. pylori infection in patients with peptic ulcer dyspepsia and non-ulcer dyspepsia. We then performed analyses stratified by the following trial characteristics: age of subjects (children vs. adult vs. elders) and treatment regimens (different PPIs).
To evaluate potential publication bias, we constructed a funnel plot for the primary outcome of eradication rates of H. pylori, using log RR as the measure of effect, and visually inspected it for asymmetry by plotting effect sizes on the vertical axis against their corresponding standard errors on the horizontal axis. The robustness of the pooled estimate was checked by influence analysis (fixed-effects model). Each study estimate was individually omitted from the data set, followed in each case by recalculation of the pooled estimate of the remaining studies (20). We also carried out Egger’s regression intercept and Begg’s rank correlation tests to assess this asymmetry formally. The analysis was done in stata software, version 9.2 (Stata Corp, College Station, TX, USA) using the ‘metan’ command. We considered a one-tailed P-value of less than 0·05 as significant.
Description of the studies
Of 145 potentially relevant studies, we excluded 29 reviews, seven comments, two letters and four editorials after reviewing titles and abstracts; then excluded 95 studies (pilot study, non-clinical studies, different interventions) after retrieving the full paper, including a post hoc analysis of a previously reported trial (21). Three abstracts from the Science Citation Index were eligible (14–16). Finally, eight published full-text manuscripts and three abstracts met our inclusion criteria for meta-analysis (6–16). The flowchart of reviews showed the detailed process of selection (Fig. 1). Two articles need to be emphasized. The study by Zhao et al. (22) was excluded because they considered furanzolidone instead of clarithromycin in the sequential regimen and that by Francavilla et al. (9) was included according to the North American Society Pediatric Gastroenterology, Hepatology and Nutrition guideline, although they considered metronidazole instead of clarithromycin in the triple therapy.
A total of 2883 participants (1338 for sequential regimen and 1545 for standard triple regimens) were enrolled in the 11 included studies. Characteristics of the 11 studies used in the meta-analysis are summarized in Tables 1 and 2. All of these studies were published in English except for a French study by Kalach et al. (13). There was only one trial (12), which was given in a blinded open-label fashion. Three trials were in children (9, 13, 16) and one trial in elderly patients (10). Of the 11 studies, seven trials evaluated the efficacy of 7-day triple regimen vs. sequential therapy (6, 9, 10, 13–16), two using a 10-day triple regimen vs. sequential therapy (8, 12), and two using both a 7-day triple regimen and a 10-day triple regimen vs. sequential therapy (7, 11). Table 3 shows the results of quality assessments by the Jadad scale. Weighted kappa statistic calculated between two reviewers showed a high degree of agreement (κ = 0·88).
Table 1. Characteristics of the 11 studies used in the meta-analysis
Type of publication
Age (mean ± SD)
Smoking habit (yes/no)
Test for confirming infection
Test for confirming eradication
PUD, peptic ulcer dyspepsia; NUD, non-ulcer dyspepsia; RUT, rapid urease test; UBT, urea breath test; NR, not reported; HpSA, Helicobacter pylori stool antigen test; C, clarithromycin; A, amoxicillin; O, omeprazole; R, rabeprazole; T, tinidazole; P, pantoprazole; M, metronidazole; E, esomeprazole.
Sequential therapy vs. 7-day triple therapy. Nine trials compared sequential therapy vs. 7-day triple therapy with a total of 2299 patients (6, 7, 9–11, 13–16). In intention-to-treat analysis, successful eradication rate with sequential therapy was achieved in 1069 of 1143 patients (93·5%; 95% CI 93·4–93·6) compared with 880 of 1156 patients treated with 7-day triple regimen (76·1%; 95% CI 75·8–76·4). The summary RR was 1·23 (95% CI 1·19–1·27), which demonstrated clear superiority of the sequential therapy over the 7-day triple regimen (Fig. 2). There was no significant heterogeneity between the trial results (I2 = 0·0%). Corresponding per protocol eradication rates were 95·4% (1046/1096) with sequential therapy vs. 77·8% (860/1105) for 7-day triple regimen. The summary RR was 1·23 (95% CI 1·19–1·27). To examine the influence of individual studies, we also performed an influence analysis using the metainf command. We found that all the estimates of effect were similar and in fact yielded more significant, narrower confidence limits (Fig. 3).
Sequential therapy vs. 10-day triple therapy. Head-to-head comparisons between sequential therapy and 10-day triple therapy are available in four studies (7, 8, 11, 12). The range of H. pylori eradication rates achieved was 76·7–81·7% with a mean H. pylori eradication rate of 79·2% (308/389) for 10-day triple therapy and 89·0–95·6% with a mean H. pylori eradication rate of 92·4% (354/383) for sequential therapy. The summary RR was 1·16 (95% CI 1·10–1·23) (Fig. 4).
Eradication of H. pylori infection in patients with peptic ulcer dyspepsia and non-ulcer dyspepsia
Three RCTs reported H. pylori eradication rates for patients with peptic ulcer dyspepsia and non-ulcer dyspepsia with sequential therapy and standard triple therapies (6–8). For patients with peptic ulcer dyspepsia, the overall eradication rates were 97·7% (168 of 172) in the sequential therapy vs. 81·2% (234 of 190) in 7-day or 10-day triple regimens. In intention-to-treat analysis, pooled sequential therapy achieved better results than 7- or 10-day triple therapies, the summary RR was 1·24 (95% CI 1·15–1·34) (Fig. 5). However, there was significant heterogeneity (I2 = 77·1%, P = 0·013). Heterogeneity disappeared when we included 7-day triple therapy (6, 7) or 10-day triple therapy (7, 8) independently. Pooled sequential therapy achieved better results than 7/10-day triple therapy, the summary RR was 1·27 (95% CI 1·17–1·38) and 1·06 (95% CI 0·96–1·17), respectively.
For patients with non-ulcer dyspepsia, the pooled cure rates were 91·6% (466 of 509 patients) with sequential therapy and 72·9% (416 of 571 patients) with standard triple therapies. Pooled analysis demonstrated clear superiority of the sequential therapy over triple regimen with RR was 1·26 (95% CI 1·19–1·33). There was no significant heterogeneity between trial results (I2 = 0·0%) (Fig. 5).
Eradication of H. pylori infection in patients with antibiotics resistance
Several factors have been reported to cause the failure of standard triple therapies. Among these factors, bacterial resistance to antibiotics is one of the major causes. In this review, we also pooled available data about clarithromycin resistance and metronidazole resistance. Four studies provided available data on the efficacy of H. pylori eradication in patients with clarithromycin resistance (6, 12, 13, 21), three for metronidazole resistance (6, 12, 13). A post hoc analysis of a previous trial that reported related data about primary clarithromycin resistance was included (21). For patients with clarithromycin resistance (with or without metronidazole), the mean H. pylori eradication was 70·7% (95% CI 60·4–81·0%) (53 of 75) with sequential therapy and 33·8% (95% CI 22·6–45·0) (23 of 68) with standard triple therapies. For patients with metronidazole resistance (without clarithromycin resistance), sequential therapy can achieve a mean cure rate of 96·0% (72 of 75), whereas the cure rate was 67·6% (50 of 74) with standard triple therapies. The summary RRs were 2·01 (95% CI 1·35–2·98) and 2·07 (95% CI 1·30–3·31) in the presence of clarithromycin resistance and metronidazole resistance, respectively (Fig. 6).
Six of the 11 RCTs reported the frequency of adverse events with sequential therapy and standard triple therapy (6, 7, 9–12). The occurrence of adverse events ranged from 7 to 17·5%, and 6·0 to 17·1% for sequential therapy and standard triple therapies. Common adverse events reported were diarrhoea, glossitis, abdominal pain, nausea and vomiting. Adverse events occurred in 8·4% (70/837) of those receiving sequential therapy and 8·7% (73/839) of those receiving the comparison 7-day triple regimens. Meta-analysis revealed no difference in the frequency of adverse events (RR = 0·96, 95% CI 0·70–1·31). This meta-analysis did not demonstrate statistically significant heterogeneity (I2 = 0·0%) (Fig. 7).
Subgroup analysis/sensitivity analysis
Subgroup analyses for the meta-analysis were planned to explore the effect of age of subjects (adult vs. children vs. elders) and different PPIs. For sub-analysis, sequential therapy achieved higher H. pylori eradication rates than 7-day triple regimen in the children subgroup (RR = 1·18; 95% CI 1·03–1·35), and in the elder subgroup (RR = 1·18; 95% CI 1·05–1·32). From 11 identified trials, four trials employed rabeprazole (6–8, 10), four trials employed omeprazole (9, 13, 14, 16), two trials employed esomeprazole (11, 15) and one trial employed pantoprazole (12). We then performed sub-analysis depending on different PPIs. The summary RRs were 1·25 (95% CI 1·19–1·31), 1·20 (95% CI 1·12–1·28), 1·18 (95% CI 1·09–1·29) and 1·15 (95% CI 1·03–1·27) for rabeprazole, esomeprazole, omeprazole and pantoprazole, respectively.
In the sensitivity analysis, we excluded three low-quality studies (abstracts) (14–16). The eradication rates were not significantly affected (RR = 1·25; 95% CI 1·19–1·30).
For H. pylori eradication rates with sequential therapy vs. 7-day triple regimen, no evidence of publication bias was observed, as indicated by a symmetric funnel plot and a non-significant Begg test (P = 0·754) and Egger test (P = 0·477) (Fig. 8).
The 7-day PPI triple therapy is the current first choice treatment for the eradication of H. pylori, but the eradication failure rate can be more than 20% (23). Causes of treatment failure include antibiotic resistance, poor compliance, short (7–10 days) duration of therapy, drug-related side-effects, bacterial load in the stomach, CagA status, smoking habit and gastroduodenal pathology (24). To improve the efficacy of triple therapy in those areas with >15–20% primary clarithromycin resistance, the use of 14-day regimen or a 10–14 days quadruple therapy has been recently proposed in the updated European guidelines (2). Novel therapeutic approaches to cure H. pylori are still urgently needed.
The sequential regimen is a novel 10-day therapeutic approach based on a different combination of the available antibiotics, and consisting of dual therapy for 5 days and triple therapy for the remaining 5 days (5). Many trials have demonstrated that the cure rate following the 10-day sequential regimen was significantly higher than either 7- or 10-day standard triple therapies (6–16). A pooled data analysis has recently been reported on the efficacy of sequential therapy (17).
The mechanism underlying sequential therapy’s superiority over standard triple regimens remains unclear. Amoxicillin can eradicateH. pylori in about 50% of infected patients and reduce its load in the remaining cases. The reduction of the bacterial load may improve the response to the subsequent short course of triple therapy (25). Indeed, amoxicillin can destroy the wall of bacterial cells, increasing the intracellular diffusion of the macrolide, with a consequent improved outcome. Therefore, amoxicillin in the initial dual therapy phase could prevent the selection of secondary clarithromycin resistance and might increase the efficacy of clarithromycin in the second phase of treatment (26). Furthermore, the higher efficiency could also be due to the addition of a new drug, tinidazole, to the standard regimen. Tinidazole is similar to metronidazole but has a longer duration of action; it is licensed for use in H. pylori infection in the UK although is not widely used. Sequential regimen has three antibiotics and is, therefore, more comparable with quadruple therapy, which also consists of three anti-microbial agents. However, it is neither a quadruple therapy nor two concomitant triple therapies (27). In addition, it has been argued that the higher effectiveness of the sequential regimen compared with 7-day triple could also be related to its duration, rather than to the sequential administration of drugs (28). Additional studies should be performed to clarify the mechanism and generalizability of the observed effect.
In the meta-analysis, we showed that sequential therapy achieved better results than 7-day triple regimens with an estimated 17·6% higher eradication rate (RR = 1·23; 95% CI 1·19–1·28). The results of per protocol analysis did not differ substantially from those of intention-to-treat analysis (RR = 1·23; 95% CI 1·19–1·27). Our results also demonstrated significant superiority of sequential therapy over 10-day triple regimen with a RR of 1·16 (95% CI 1·10–1·23). For patients with peptic ulcer dyspepsia or non-ulcer dyspepsia, sequential therapy led to a higher eradication rate than triple regimens, with RR of 1·24 (95% CI 1·15–1·34) and 1·26 (95% CI: 1·19–1·33), respectively. That sequential treatment was equally effective in patients with peptic ulcer disease and non-ulcer dyspepsia, suggests that this scheme, unlike triple therapy, does not discriminate between CagA-positive and CagA-negative H. pylori strains (8). Adverse events were similar with sequential therapy vs. standard triple therapies. Furthermore, economic advantage is further strengthened by the consideration that a first-line therapy with the sequential regimen is 15–20% more effective than standard 7–10 days triple therapies (17).
Limitation of the present review includes: first, our inclusion criteria for testing for H. pylori infection was not strict, because H. pylori status was determined by only one routine method; second, we included three abstracts which reported inadequate outcome data. However, sensitivity analysis with their exclusion did not alter the overall conclusion.
In Italy, the sequential regimen is given equivalent status to standard 7–14 triple therapies as first-line treatment in the updated Italian guidelines on H. pylori management (29). However, sequential therapy requires further evaluation. Although the concept appears very promising and the regimen should probably replace existing triple therapies, all of the studies of this regimen come from Italy, mainly Rome and Foggia. Before we can accept the regimen for other countries, the higher efficacy of the sequential regimen should be confirmed outside Italy. We, therefore, urgently need data from other international groups to see whether they can reproduce these very good results.
Many of these published studies have substantial limitations: small sample sizes, lack of blinding and failure to evaluate resistance. Thus, more studies are needed to examine the influence of dose and treatment duration, and the need for sequential administration of the drugs, and whether extend the duration to 14 days lead to better outcomes. More high-quality trials are also required. Furthermore, head-to-head comparisons of sequential therapy with 14-day triple therapy tell us which of these regimens can eventually replace the current first-line therapies.
Sequential therapy is a promising regimen for H. pylori eradication. However, to replace the current first-line triple therapies, further validation in other geographical areas, using robust study designs are required.