Clinical trial: clarithromycin vs. levofloxacin in first-line triple and sequential regimens for Helicobacter pylori eradication


Dr J. Molina-Infante, Department of Gastroenterology, Hospital San Pedro de Alcantara, C/Pablo Naranjo s/n, 10003 Cáceres, Spain.


Aliment Pharmacol Ther31, 1077–1084


Background Helicobacter pylori eradication rates with standard triple therapy have declined to unacceptable levels.

Aim  To compare clarithromycin and levofloxacin in triple and sequential first-line regimens.

Methods  A total of 460 patients were randomized into four 10-day therapeutic schemes (115 patients per group): (i) standard OCA, omeprazole, clarithromycin and amoxicillin; (ii) triple OLA, omeprazole, levofloxacin and amoxicillin; (iii) sequential OACM, omeprazole plus amoxicillin for 5 days, followed by omeprazole plus clarithromycin plus metronidazole for 5 days; and (iv) modified sequential OALM, using levofloxacin instead of clarithromycin. Eradication was confirmed by 13C-urea breath test. Adverse effects and compliance were assessed by a questionnaire.

Results  Per protocol cure rates were: OCA (66%; 95% CI: 57–74%), OLA (82.6%; 75–89%), OACM (80.8%; 73–88%) and OALM (85.2%; 78–91%). Intention-to-treat cure rates were: OCA (64%; 55–73%), OLA (80.8%; 73–88%), OACM (76.5%; 69–85%) and OALM (82.5%; 75–89%). Eradication rates were lower with OCA than with all the other regimens (P < 0.05). No differences in compliance or adverse effects were demonstrated among treatments.

Conclusions  Levofloxacin-based and sequential therapy are superior to standard triple scheme as first-line regimens in a setting with high clarithromycin resistance. However, all of these therapies still have a 20% failure rate.


Evolving research has demonstrated the relationship of Helicobacter pylori infection with chronic gastritis, peptic ulcer disease and gastric adenocarcinoma and MALT lymphoma, as well as the importance of a prompt cure of the infection to change the natural history of these diseases. After the initial high efficacy (eradication rate >90%) of triple standard regimens, we are witnessing within the last decade, a progressive decline in cure rates.1, 2 The high prevalence of antimicrobial drug resistance, especially to clarithromycin and metronidazole, is believed to be the key factor. For this reason, consensus statements recommend empirical therapeutic regimens that achieve H. pylori cure rates higher than 80% on an intention-to-treat (ITT) basis.3, 4

Novel antibiotic regimens have been developed to overcome this troublesome scenario. One recent therapeutic innovation, postulated as an alternative to standard triple therapy, is the so called ‘sequential’ treatment.5 Strictly speaking, it is not a new approach, as it uses well-known drugs with approved indication for H. pylori eradication. However, the administration strategy is innovative. The sequential regimen is a simple dual therapy including a proton pump inhibitor (PPI) plus amoxicillin 1 g (both twice daily) given for the first 5 days followed by a triple therapy including a PPI, clarithromycin 500 mg and tinidazole (all twice daily) for the remaining 5 days. Its rationale is based on an initial phase with amoxicillin, which aims to lower the bacterial load in the stomach. Moreover, it has been speculated that amoxicillin may prevent the development of efflux channels for clarithromycin. This induction phase therefore is believed to amplify the efficacy of the second phase of therapy containing clarithromycin and metronidazole. Italian studies regarding this clarithromycin-based sequential therapy have shown promising eradication rates higher than 90%, even in patients with risk factors for triple therapy failure (clarithromycin resistance, non-ulcer dyspepsia, smoking or the absence of the gene CagA).6, 7 Furthermore, a recent meta-analysis has shown that eradication rate with 10-day sequential therapy (93.4%) is notably higher than that for standard triple therapy (76.9%), with similar adherence in both groups.8 Thus, it has been questioned whether sequential therapy should be the preferred first line therapy for H. pylori infection9 albeit the global validation of the sequential scheme outside Italy is awaited.

On the other hand, levofloxacin, a fluoroquinolone with in vivo activity against H. pylori strains resistant to clarithromycin and metronidazole, has also shown promising results in different first-line triple regimens in Italy, Spain and the Netherlands, with an eradication rate on ITT ranging from 83% to 96%.10–17 The efficacy of levofloxacin in a sequential eradication scheme for H. pylori infection has been exclusively assessed in a single recent study from Turkey,18 with an 82% ITT cure rate. Moreover, a head-to-head comparison between similar clarithromycin and levofloxacin regimens has not been addressed yet. Therefore, we aimed to evaluate the cure rate of triple and sequential regimens containing clarithromycin or levofloxacin in a geographical area with a high failure rate of triple classical eradication therapy.


This study is a prospective, open-label, single centre, randomized trial. From January 2008 to August 2009, 460 consecutive H. pylori positive patients were enrolled. The diagnosis of H. pylori was determined by at least a positive test among urea breath test, histology or rapid urease test. Written informed consent was obtained from all patients. The study was approved by the Ethics Committee of our Hospital. Exclusion criteria were (i) age under 18 years, (ii) presence of severe comorbidities, (iii) prior H. pylori eradication, (iv) gastric surgery, (v) allergy of any of the antibiotics used in the study and (vi) intake of antibiotics, PPIs or nonsteroidal anti-inflammatory drugs within the last month.

Using a computer-generated numeric sequence, patients were randomized to receive one of the four first-line schemes for 10 days, having all high-dose acid suppressive therapy:19

  • Standard triple OCA (n = 115), omeprazole 20 mg, clarithromycin 500 mg and amoxicillin 1 g, all three twice a day.

  • Triple OLA (n = 115), omeprazole 20 mg, levofloxacin 500 mg and amoxicillin 1 g, all three twice a day.

  • Sequential OACM (n = 115), omeprazole 20 mg and amoxicillin 1 g both twice a day for 5 days followed by omeprazole 20 mg, clarithromycin 500 mg and metronidazole 500 mg, all three twice a day during the next 5 days.

  • Modified sequential OALM (n = 115), omeprazole 20 mg and amoxicillin 1 g both twice a day for 5 days followed by omeprazole 20 mg, levofloxacin 500 mg and metronidazole 500 mg, all three twice a day during the next 5 days.

Compliance with therapy was determined from the interrogatory and the recovery of empty envelopes of medications. Side effects were assessed by a specific questionnaire completed at the time of post-treatment urea breath testing. Eradication of H. pylori infection was defined as a negative urea breath test 8 weeks after completion of treatment except for patients requiring a follow-up endoscopy because of gastric ulcer, in which histological examination of four samples taken from the body and the antrum stained with Giemsa was the diagnostic test.

Statistical analysis

The predetermination of the sample size in each group (n = 115), assuming an 80% power at a 5% significance level, was performed to detect a 15% difference in eradication rates between triple standard therapy (control group) and any of the other regimens (experimental group), in accordance with the most recent cure rates reported in our setting (triple standard therapy 70%,20 sequential therapy 84%).21

Categorical variables are described with percentages and continuous variables are described with mean and standard deviation or median and range as appropriate. The eradication rates and their 95% confidence intervals (95% CI) were obtained by ITT and per protocol (PP). Univariate and multivariate logistic regression analyses were performed to evaluate independent predictive variables for eradication of H. pylori. The magnitude of the effect is described with the odds ratios (ORs) and 95% CI. Variables chosen to be introduced in predictive models depended on statistical significance on univariate analysis. P-values lower than 0.05 were considered statistically significant.


Four hundred and sixty patients were enrolled in the study. The baseline demographic and clinical characteristics, indication for eradication and diagnostic method for the infection of the total cohort and each therapeutic subgroup are listed in Table 1. The overall sample was balanced in gender, had a median age of 48 years and 33% smoked cigarettes. Dyspepsia was the most common indication for H. pylori eradication (59%).

Table 1.   Demographic and clinical characteristic of patients from the total cohort and in each therapeutic group after randomization
  1. OCA, triple standard; OLA, levofloxacin triple therapy; OACM, sequential therapy with claritrhomycin; OALM, modified sequential therapy with levofloxacin.

No. patients460115115115115
Gender (M/F), %47/5347/5353/4747/5340/60
Age, years48 (18–84)44 (19–78)51 (18–84)49 (18–80)49 (19–79)
Smoking habit (%)3335333829
Indication, n (%)
 Non-investigated dyspepsia95 (20.7)30 (26.1)23 (20)25 (21.7)20 (17.4)
 Functional dyspepsia180 (39.1)42 (36.6)47 (40.8)42 (36.5)51 (44.3)
 Gastric ulcer86 (18.7)18 (15.6)20 (17.4)24 (20.8)23 (20)
Duodenal ulcer78 (17)18 (15.6)22 (19.1)20 (17.4)17 (14.8)
 Gastric cancer in first-degree relatives21 (4.5)7 (6)3 (2.6)4 (3.5)4 (3.5)
Diagnostic method, n (%)
 Urea breath test165 (35.8)41 (35.6)41 (35.6)39 (33.9)31 (26.9)
 Rapid urease test121 (26.3)29 (25.2)32 (27.8)27 (23.442 (36.5)
 Histology174 (37.8)45 (39.1)42 (36.5)49 (42.6)42 (36.5)

The study flow chart is summarized in Figure 1. Eradication rates are shown in Figure 2. ITT cure rates were: OCA (64%; 55–73%), OLA (80%; 73–88%), OACM (76%; 69–85%) and OALM (82%; 75–89%). PP cure rates were: OCA (66%; 95% CI: 57–74%), OLA (82%; 75–89%), OACM (80%; 73–88%) and OALM (85%; 78–91%). ITT eradication rates were significantly higher with all the other regimens when compared to standard triple treatment [OLA OR (95% CI): 2.4 (1.3–4.5), P < 0.05, OACM OR (95% CI): 2.1 (1.1–3.9), P < 0.05, OALM OR (95% CI): 2.9 (1.5–5.6), P < 0.05]. No significant differences were demonstrated among OLA, OACM and OALM. Figure 3 shows the ITT eradication rates according to the indication. In patients with non-investigated dyspepsia, no relevant differences were observed among the four therapeutic regimens. Levofloxacin-based regimens were significantly more effective for patients suffering from functional dyspepsia when compared with OCA [OLA OR (95% CI): 4.7 (1.7–13), P < 0.01, OALM OR (95% CI): 3.5 (1.3–9.2), P < 0.05]. Clarithromycin sequential regimen failed to improve significantly eradication rates compared with OCA in this subset of patients [OR (95% CI): 1.8 (0.7–4.5)]. For patients with gastric or duodenal ulcer, OALM regimen was also significantly better than OCA [OR (95% CI): 4.1 (1.1.–14)].

Figure 1.

 Study flow chart. ITT, intention-to-treat; OCA, triple standard; OLA, levofloxacin triple therapy; OACM, sequential therapy with claritrhomycin; OALM, sequential therapy with levofloxacin.

Figure 2.

 Eradication rates per protocol and per intention-to-treat analysis for 10-day clarythromycin and levofloxacin regimens. * P < 0.05. OCA, triple standard; OLA, levofloxacin triple therapy; OACM, sequential therapy with claritrhomycin; OALM, modified sequential therapy with levofloxacin.

Figure 3.

 Eradication rates in the intention-to-treat analysis depending on the indication for eradication among all the regimens. * P < 0.05. OCA, triple standard; OLA, levofloxacin triple therapy; OACM, sequential therapy with claritrhomycin; OALM, modified sequential therapy with levofloxacin.

In the univariate analysis, the efficacy of the therapy was influenced by the type of treatment (OCA 66% vs. 82% for the others, P < 0.01), but not by the indication (77% for dyspepsia vs. 82% for ulcer, P = 0.18), gender (82% for men vs. 75% for women, P = 0.06), age [eradicated 48 years (18–87) vs. non-eradicated 49 years (18–71), P = 0.4] or smoking habit (smokers 81% vs. nonsmokers 77%, P = 0.28). In the multivariate analysis, any of the therapeutic schemes different from OCA remained as independent predictor factor for H. pylori eradication [OR (95% CI): 2.5 (1.5–4)].

Overall, 97% of the patients had complete adherence to antibiotic therapy as shown in Figure 1. Three patients were lost to follow-up and nine patients discontinued therapy because of adverse events. Minor or mid side effects were reported by 129 patients (28%), without significant differences between therapeutic schemes. No major adverse events were observed. Side effects are summarized in Table 2.

Table 2.   Adverse events resulting from antibiotic therapy in patients from the total cohort and in each therapeutic group after randomization, n (%)
  1. OCA, triple standard; OLA, levofloxacin triple therapy; OACM, sequential therapy with claritrhomycin; OALM, modified sequential therapy with levofloxacin.

Side effects129 (28)29 (25)32 (27)29 (25)29 (25)
Diarrhoea38 (8.2)1012610
Metallic taste30 (6.5)7472
Epigastralgia/nausea24 (5.2)7386
Myalgias17 (3.6)0809
Aphthous stomatitis12 (2.6)2352
Oral candidiasis4 (0.8)2200
Skin rash3 (0.6)1020
Light-headedness1 (0.02)0010


This is the first randomized trial that compares clarithromycin to levofloxacin in triple and sequential schemes for H. pylori eradication. As several recent meta-analyses8, 22, 23 have shown the advantage of sequential therapy over the standard triple scheme for clarithromycin-resistant strains, it has been suggested that sequential therapy should be advisable as a first-line therapy when the prevalence of clarithromycin resistance is high, as it occurs presently in most developed countries. The present study shows that the eradication rate for clarithromycin sequential therapy was suboptimal (76%) in a setting with a high rate of failure for triple standard therapy (64%). Indeed, this is the lowest eradication rate reported to date for sequential therapy. Therefore, the sequential regimen may not lead to acceptable eradication rates in areas with a high prevalence of clarithromycin resistance. Indeed, ITT eradication rates for standard triple therapy (75–79%), in the aforementioned Italian studies, may be an indirect marker of medium clarithromycin resistance.7, 8 Moreover, sequential therapy achieved a 75% eradication rate (41/55 patients) when exposed to documented clarithromycin-resistant strains in literature, whereas it was absolutely ineffective (0% eradication rate) in patients with dual resistance to clarithromycin and metronidazole.5

In this regard, the validation process of the sequential therapy regimen outside Italy has led to controversial results. Similar successful eradication rates have been obtained in other settings such as in Thailand (96%)24 or in Taiwan (92%),25 whereas much more modest results have been recently obtained in Panama (85%),26 France (85%),27 Spain (84%)21 and Korea (80% and 77.9%).28, 29 In fact, most of the studies regarding sequential therapy published during 2008 and 2009 had eradication rates lower than 90% and, in some cases, even ≤80%.24–28 Further controversy regarding sequential therapy has been provoked with the results of several studies performed outside Italy that have been unable to demonstrate differences between sequential and standard triple regimens.26–29 Thus, studies validating both the cure rate of sequential therapy and its advantage over triple standard therapy in settings with different patterns of antimicrobial resistance are required before it can be widely recommended in clinical practice. Attending to the results of the present study, it will be especially mandatory in geographical areas with high rate of clarithromycin-resistant H. pylori strains.

Alternative therapeutic regimens are awaited in settings where empirical therapies do not produce a ≥80% cure rate on ITT basis. Besides the use of levofloxacin-triple therapy in second and third-line regimens, encouraging cure rates (≥90%) have been obtained in first-line schemes in Italy10, 11, 13, 14 and the Netherlands.17 More recently, a novel levofloxacin modified sequential therapy has been described for the first time in Turkey with cure rates of 86% on PP and 82.5% on ITT.18 In the present study, levofloxacin-based triple (80%) and modified sequential scheme (82%) achieved in the present study ‘adequate’ cure rates. These results for the modified sequential regimen are identical to those of the Turkish study. However, the overall results for both levofloxacin regimens in the present study are to be considered poor. A first course of antibiotic therapy for H. pylori infection should only be considered good if it achieves a cure rate on ITT basis >90%.30 Indeed, fluoroquinolones resistance has been rapidly increasing and this antibiotic family will probably become useless in the short term.20, 30 This trend is confirmed by our results using levofloxacin triple regimen in first-line therapy for H. pylori since 2007 (84.4%,15 82.7%16 and 80.8% in the present study). Moreover, another recent Spanish study31 could not demonstrate differences between standard triple therapy (75%) and levofloxacin triple therapy (72%) on an ITT basis. Similar to our results with sequential therapy, the collective Spanish results represent the lowest eradication rates reported for triple levofloxacin regimen in first-line therapy, which contrast with cure rates ≥90% reported in other settings.

Therefore, the current prevalence of antibiotic resistance (clarithromycin, metronidazole and possibly levofloxacin) has probably increased to such an extent in some settings, such as Spain, that all patients should be considered as having resistant H. pylori infections to maintain acceptable cure rates.30 In this troublesome scenario, alternative regimens for initial treatment merit further evaluation, such as traditional bismuth-based quadruple therapy and concomitant therapy (nonbismuth-containing therapy giving the four constituent components of sequential therapy concurrently for 5–10 days).5, 20, 30 As bismuth-based quadruple therapy is not influenced by metronidazole or macrolide resistance, it has been recommended as the treatment of choice when clarithromycin resistance rates are ≥15% in the community.32, 33 Nevertheless, a recent meta-analysis failed to find a significant difference in the success rate in primary treatment between quadruple and standard triple therapy,34 which were both suboptimal (<80%). Additional disadvantages of bismuth quadruple therapy are that it is the most complex therapy, even despite a novel single capsule containing bismuth, metronidazole and tetracycline and that bismuth salts are not available anymore in many countries.5, 32

On the other hand, the advantage of concomitant therapy over triple therapy has been recently demonstrated, as well as its equal effectiveness and safety with less complexity compared with sequential therapy.25, 35 The first randomized trial comparing concomitant and sequential therapies concluded that eradication rate was significantly better with concomitant therapy for clarithromycin-resistant strains (75% vs. 57%) and dual clarithromycin and metronidazole resistant-strains (75% vs. 33%).25 Overall, clarithromycin-based sequential and concomitant therapies are encouraging novel regimens for primary H. pylori eradication, but it remains to be elucidated whether both schemes are really effective in settings with a high prevalence of clarithromycin resistance, as it occurs presently in most developed countries.

The present study has several shortcomings. A major drawback is the lack of pre-treatment susceptibility testing for clarithromycin and levofloxacin. Prevalence of clarithromycin resistance has been reported to be <15% in adults in our country.36, 37 Nonetheless, more recent studies have highlighted an increasing trend in its prevalence in adults and children (22%),38 with primary H. pylori resistant-strains in children as high as 49% for clarithromycin and 32% for metronidazole.39 On the contrary, levofloxacin seems to have a relatively low resistance rate (<6%) in our country,38, 40 which contrasts with the increasing trend in other settings, ranging from 14% to 22%.41 However, the higher cost of levofloxacin regimens is a major concern when dealing with a common disease like H. pylori infection. As such, it is advisable to currently save levofloxacin for second and third line regimens.

In conclusion, the efficacy of sequential therapy may be suboptimal (<80%) in areas with a high rate of failure for triple standard therapy. Sequential therapy is a novel promising approach that deserves consideration, but its cure rates and its advantage over standard triple regimen require validation outside Italy before a generalized change is recommended in clinical practice, especially in settings with high rate of clarithromycin-resistant H. pylori strains. Currently, levofloxacin triple and modified sequential therapies are good alternatives, albeit burdened with a 20% failure rate and a higher cost. On this point, further studies addressing and comparing both sequential and concomitant quadruple schemes are warranted.


Declaration of personal interests: We are indebted to the hard work and dedication of Juana Vicenta Molano, the nurse who performed the post-therapy assessment. Declaration of funding interests: CIBEREHD is funded by the Instituto de Salud Carlos III.