Background and aims Antimicrobial drug resistance is a major cause of the failure of Helicobacter pylori eradication and is largely responsible for the decline in eradication rate. Quadruple therapy has been suggested as a first-line regimen in areas with clarithromycin resistance rate >15%. This randomised trial aimed at evaluating the efficacy of a levofloxacin-containing sequential regimen in the eradication of H pylori-infected patients in a geographical area with >15% prevalence of clarithromycin resistance versus a clarithromycin-containing sequential therapy.
Methods 375 patients who were infected with H pylori and naïve to treatment were randomly assigned to one of the following treatments: (1) 5 days omeprazole 20 mg twice daily + amoxicillin 1 g twice daily followed by 5 days omeprazole 20 mg twice daily + clarithromycin 500 mg twice daily + tinidazole 500 mg twice daily; or (2) omeprazole 20 mg twice daily + amoxicillin 1 g twice daily followed by omeprazole 20 mg twice daily + levofloxacin 250 mg twice daily + tinidazole 500 mg twice daily; or (3) omeprazole 20 mg twice daily + amoxicillin 1 g twice daily followed by omeprazole 20 mg twice daily + levofloxacin 500 mg twice daily + tinidazole 500 mg twice daily. Antimicrobial resistance was assessed by the E-test. Efficacy, adverse events and costs were determined for each group.
Results Eradication rates in the intention-to-treat analyses were 80.8% (95% CI, 72.8% to 87.3%) with clarithromycin sequential therapy, 96.0% (95% CI, 90.9% to 98.7%) with levofloxacin-250 sequential therapy, and 96.8% (95% CI, 92.0% to 99.1%) with levofloxacin-500 sequential therapy. No differences in prevalence of antimicrobial resistance or incidence of adverse events were observed between groups. Levofloxacin-250 therapy was cost-saving compared with clarithromycin sequential therapy.
Conclusion In an area with >15% prevalence of clarithromycin resistant H pylori strains, a levofloxacin-containing sequential therapy is more effective, equally safe and cost-saving compared to a clarithromycin-containing sequential therapy.
- H pylori, sequential therapy
Statistics from Altmetric.com
Significance of this study
What is already known about this subject?
The prevalence of H pylori strains resistant to commonly used antimicrobials is increasing and this represents the major cause of treatment failures.
The Maastricht III Consensus has established that clarithromycin should not be used in areas with higher than 15% H pylori clarithromycin-resistant strains.
Sequential therapy is a promising alternative for eradication of H pylori infection.
Fluoroquinolones such as levofloxacin have been used successfully in combination with PPI and amoxicillin in the treatment of H pylori infection.
What are the new findings?
Levofloxacin-containing sequential therapy in a setting of high prevalence of clarithromycin-resistant H pylori strains is highly effective, reaching eradication rates higher than 95%.
Levofloxacin-containing sequential therapy is cost-saving and well tolerated.
Clarithromycin-containing sequential therapy does not achieve satisfactory eradication rates in a geographical area with high (ie, >15%) prevalence of H pylori strains resistant to clarithromycin or with dual (ie, clarithromycin + metronidazole) resistance.
How might it impact on clinical practice in the foreseeable future?
The result from this clinical trial strongly suggests that levofloxacin-containing sequential therapy should be used as first-line eradication regimen in areas with high prevalence of clarithromycin and/or clarithromycin + metronidazole resistant strains of H pylori.
Helicobacter pylori infection is the main cause of chronic gastritis, peptic ulcer, adenocarcinoma of the distal stomach and gastric mucosa associated lymphoid tissue lymphoma.1 2 Therefore, a safe and effective eradication regimen for this infection is imperative.3 While the prevalence of H pylori infection has been falling in developed countries, difficulties with eradication have been increasing as the prevalence of H pylori strains resistant to the most commonly used antimicrobials increases.4–6 As a result, current recommended first-line therapies both in the US and Europe relying on clarithromycin (CLA) or metronidazole (MET) in conjunction with amoxicillin (AMO) and proton pump inhibitors (PPI) achieve an approximately 75–80% eradication rate, which is far from being acceptable.7 8
CLA resistance is considered crucial to the success of a therapeutic regimen, because, unlike MET, CLA resistance cannot be overcome by increasing the dose of the drug. The increase in H pylori strains with primary resistance to CLA is becoming a major problem in many Western countries and reaches as high as 13% in the US and 24% in some European countries.9 10 Because of this, the Maastricht III Consensus suggested that CLA should not be used in areas with a CLA resistance rate >15% and indicated bismuth-containing quadruple therapy as the first-line H pylori eradication regimen in this setting.11
Recently, a novel 10-day sequential therapy (ST) consisting of a 5-day dual therapy (PPI standard dose, twice daily + AMO 1 g, twice daily) followed by a 5-day triple therapy (PPI, standard dose, twice daily + CLA 500 mg, twice daily + tinidazole (TIN) 500 mg, twice daily) has shown good eradication rates both in elderly and paediatric patients.12–14 A meta-analysis of 10 randomised clinical trials involving 2747 patients has indicated that ST achieves eradication rates significantly higher than standard triple even when there was evidence of CLA resistance.15 However, the eradication rate achieved with ST should still be improved.16
We have recently shown that, in our region of southern Italy, the prevalence of CLA-resistant H pylori strains was approximately 18% in patients naïve to treatment and reached up to 45% in patients with multiple unsuccessful eradication therapies.17 18 On the other hand, primary resistance to levofloxacin (LEV) was about 3%.18 We have therefore hypothesised that a LEV-containing ST (LEV-ST) might be superior to CLA-containing ST (CLA-ST) in this clinical setting and designed a study with the primary endpoint of assessing whether LEV-ST was more efficient than CLA-ST in eradication of H pylori infection in an area with high prevalence of CLA-resistant H pylori strains. Secondary endpoints were to assess the influence of anti-microbial resistance on the outcome of eradication treatments, the incidence of adverse events, and the cost-effectiveness of either regimen.
Patients and methods
This was a prospective, randomised, controlled, multicentre study with a parallel-group design. At baseline, patients were evaluated for inclusion and exclusion criteria and provided written informed consent. Patients were then randomly assigned to a treatment group and had a follow-up evaluation to assess the eradication rate of H pylori infection and occurrence of adverse events. The study was performed according to good clinical practice and the Declaration of Helsinki, and was approved by the institutional Ethical Committee. All patients in whom eradication of the infection failed were offered a second gastroenterology consultation, rescue therapy and re-testing.
Nine hundred and eighty-five consecutive patients with dyspepsia who had been referred to our Gastroenterology Units for consultation between May 2008 and October 2009 were asked to participate in the study. H pylori-infected patients who were at least 18 years of age and who had never received H pylori eradication treatment were included in the study. Diagnosis of H pylori infection was based on positivity to a 13C urea breath test (13C UBT) or by positivity to both a rapid urease test and histology in those patients who underwent endoscopy because they were more than 45 years of age and/or because of the presence of alarm symptoms. In patients undergoing endoscopy, five biopsy (two antrum, two body and one angulus) specimens were taken for histological assessment according to the Sydney system and two more specimens (one from antrum and one from body) were taken for a rapid urease test. We did not perform pre-treatment susceptibility testing because of the ethical concern that it was not acceptable to treat patients with an empirical therapy while having the results of in vitro anti-microbial susceptibility unless we were testing efficacy against resistance. In this respect, we have in fact previously demonstrated that pre-treatment anti-microbial susceptibility testing significantly improves the efficacy of eradication therapy compared with empirical therapy.17 However, because our laboratory is highly involved in monitoring H pylori anti-microbial resistance,17–19 we performed post-treatment susceptibility testing on frozen samples obtained from those patients who had undergone endoscopy (table 1). Two additional biopsy samples were obtained from antrum for bacterial culture and anti-microbial susceptibility testing. Biopsy samples were sent to our microbiology laboratory within 24 h and were stored at −70°C. Isolated strains were tested for in vitro susceptibility to CLA, LEV, MET, AMO, tetracycline (TET) by the E-test as described previously.17–19 Exclusion criteria were: previous treatment for H pylori infection, use of inhibitors of acid secretion and/or antibiotics during the 6 weeks before the study, gastrointestinal malignancy, previous gastro-oesophageal surgery, severe concomitant cardiovascular, respiratory or endocrine diseases, clinically significant renal or hepatic disease, haematological disorders, any other clinically significant medical condition that could increase risk, history of allergy to any of the drug used in the study, pregnancy or lactation, alcohol abuse, drug addiction, severe neurological or psychiatric disorders, and long-term use of corticosteroids or anti-inflammatory drugs.
Patients were enrolled by the medical personnel of the gastrointestinal unit after assessment of appropriate indication and ruling out any contraindication to the treatments. Patients were randomly allocated to receive one of the three schedules of treatment using a centralised computer-generated random number list with block size of 3, 6, 9 and 12. An independent medical staff member assigned subjects to the three schedules. In each block there were serially numbered, sealed, opaque envelopes. Each patient received the next pack stored in the centre following ascending order of labels.
Patients were interviewed at completion of therapy to assess adherence to the therapeutic regimen and adverse events by medical personnel blinded to the eradication regimen of each patient. In particular, first, open-ended questions on side effects and then specific questions on anticipated adverse events were asked. Six and 10 weeks after completion of therapy, H pylori status was re-evaluated by 13C UBT by non-medical personnel unaware of the eradication regimen of each patient to assess whether the infection had been successfully eradicated. Infection was considered eradicated if patients tested negative on both occasions. 13C UBT was performed after an overnight fast. A baseline breath sample was obtained, and 100 mg of 13C urea with citric acid (1.4 g) was administered as an aqueous solution (Expirobacter, SOFAR, Milan, Italy). Another breath sample was collected 30 min later. The test was considered positive if the difference between the baseline sample and the 30-min sample exceeded 5.0 parts/1000 of 13CO2. All breath tests were analysed at the same laboratory by one of us (AR) by using a single gas isotope ratio mass spectrometer (ABCA, Europe Scientific, Crewe, UK).
Each group of patients consisted of 125 patients. One group was treated with CLA-ST (ie, OME 20 mg twice daily + AMO 1 g twice daily for 5 days followed by OME 20 mg twice daily + CLA 500 mg twice daily + TIN 500 mg twice daily for five more days). The second group was treated with LEV250-ST (ie, OME 20 mg twice daily + AMO 1g twice daily for 5 days followed by OME 20 mg twice daily + LEV 250 mg twice daily + TIN 500 mg twice daily for five more days), and a third group with LEV500-ST (ie, OME 20 mg twice daily + AMO 1g twice daily for 5 days followed by OME 20 mg twice daily + LEV 500 mg twice daily + TIN 500 mg twice daily for five more days). Patients who failed to eradicate the infection were re-treated with quadruple therapy consisting of OME 20 mg twice daily + MET 500 mg three time a day + TET 500 mg four times a day + bismuth subcitrate 240 mg twice daily, and after 6 weeks underwent a new 13C UBT.
Measurements and outcomes
The primary outcome of the study was eradication of H pylori infection with CLA-ST and LEV-STs. Secondary outcomes were to evaluate any difference in the efficacy of two different doses of LEV (ie, 250 mg twice daily vs 500 mg twice daily), to assess the frequency of adverse events, and to determine the cost-effectiveness of different eradication regimens.
Both intention-to-treat (ITT) and per-protocol (PP) analyses were used for the assessment of the eradication rates of H pylori infection in the three groups. Differences between frequency were evaluated by χ2 test. The significance level was set at p<0.05. The 95% CIs were constructed by normal approximation. We calculated a sample size of 125 to detect a 15% difference in the eradication rate between the CLA-ST (assumed to have and eradication rate of 80–85%) and LEV-ST (estimated to have an eradication rate of 95–100%) with a power of 0.80 and a significance level of 0.05 (one-sided α level=0.05). Assumption regarding the eradication rates of CLA-ST and LEV-ST is based on data from literature and from our preliminary observations.20–23
Cost analysis was performed using the ICERconf.EXE software. The cost-effectiveness ratio and incremental cost-effectiveness ratio24 were calculated. The costs per patient assessment were the cost of office visits, 13C UBT, and cost per tablet of OME 20 mg, AMO 1 g, CLA 500 mg, LEV 500 mg, LEV 250 mg, TIN 500 mg, MET 250 mg, TET 250 mg, bismuth subcitrate 120 mg. Cost estimates for procedures and physician fees were based on the Italian National Health System reimbursement scheme. The incremental cost of the new (ie, LEV-containing) strategy was compared with the standard regimen (ie, CLA-containing). As the economic analysis was conducted from the perspective of the prescriber, only the direct costs of treatment were included in the analysis.
Figure 1 shows the flow of patients through the study. Of the 985 patients who participated in the first visit, 583 were excluded because they were not infected with H pylori. Of the remaining 402 patients, five were excluded because of gastric cancer, two were excluded because of gastric mucosa associated lymphoid tissue lymphoma, and 20 withdrew consent. Of the 375 patients randomly assigned, three in the LEV250-ST group, two in the LEV500-ST group and two in the CLA-ST group were lost at follow-up. One patient in the CLA-ST group discontinued therapy after 3 days because of severe diarrhoea and epigastric pain. Therefore, 375 patients were included in the ITT analysis and 367 in the per protocol PP analysis. The demographic and clinical characteristics of patients were comparable in the three groups (table 1).
Eradication of H pylori infection and adverse events
The efficacy of the three eradication regimens is indicated in figure 2. In particular, in the PP analysis, eradication rates were 101/122=82.8% (95% CI, 74.9% to 89.0%) with CLA-ST, 120/122=98.3% (95% CI, 94.2% to 99.8%) with LEV250-ST, and 121/123=98.4% (95% CI, 94.2% to 99.7%) with LEV500-ST, LEV250-ST and LEV500-ST were significantly more effective than CLA-ST (p<0.0001). In the ITT analysis, eradication rates were 101/125=80.8% (95% CI, 72.8% to 87.3%) with CLA-ST, 120/125=96.0% (95% CI, 90.9% to 98.7%) with LEV250-ST, and 121/125=96.8% (95% CI, 92.0% to 99.1%) with LEV500-ST. Once again, LEV250-ST and LEV500-ST were associated with significantly higher eradication rates than CLA-ST (p<0.0001). No significant differences were observed between LEV250-ST and LEV500-ST. Patients with unsuccessful first-line therapy were treated with bismuth-containing quadruple therapy which led to eradication of the infection in the four patients who had failed to eradicate the infection with LEV-STs and in 9/11 patients in whom the infection had not been eradicated following CLA-ST (data not shown).
The three eradication regimens were well tolerated and only one patient who had been randomly assigned to CLA-ST group discontinued treatment because of severe diarrhoea and epigastric pain after 3 days of therapy. Twenty-nine patients in the CLA-ST group, 27 in the LEV250-ST group, and 29 in the LEV500-ST group reported side effects that were not sufficiently severe to cause discontinuation of therapy. Only one patient in the LEV500-ST group reported muscle pain during the second period of the treatment. The most frequently reported adverse events were diarrhoea, epigastric pain, and bad taste/taste alteration. The latter occurred more frequently during the second period of the treatment independently of the eradication regimen and was referred by all patients to the ingestion of tinidazole. The adverse events reported by patients are summarised in table 2.
Role of primary resistance in the eradication of H pylori infection
Bacterial culture was performed in those patients who had undergone endoscopy and was successful in approximately 86% of patients. Data of anti-microbial resistance were therefore available in 55/60 (91.6%), 69/81 (85.2%), 65/78 (83.3%), subjects in the CLA-ST, LEV250-ST and LEV500-ST groups, respectively. No resistance to AMO and TET was detected in any patients (data not shown). The overall prevalence of resistance to CLA, MET, LEV and CLA+MET was 39/189 (20.6%), 50/189 (26.4%), 7/189 (3.7%), and 11/189 (5.8%), respectively, and no significant differences in anti-microbial resistance were observed between groups (table 3).
CLA-ST eradicated the infection in 9/12 (75.0%) patients with isolated resistance to CLA and in 13/14 (92.8%) patients with isolated MET resistance whereas LEV-STs were both 100% effective in the CLA-resistant and MET-resistant patients (table 4). Three patients infected with dual resistant (ie, CLA+MET) H pylori strains failed to eradicate the infection with CLA-ST, whereas eight patients with dual resistant strains were successful eradicated with LEV-STs (table 4). In patients infected with LEV-resistant H pylori strains, CLA-ST was effective in 1/2 (50%), LEV250-ST in 1/2 (50%), and LEV500-ST in 2/3 (75%) patients (table 4).
The cost of 10-day ST with CLA was €65.33/patient (including the costs of drugs plus office visit and 13C UBT) compared with €84.01/patient with LEV250-ST and €100.87/patient with LEV500-ST (data not shown). Treatment failures led to a supplementary office visit, a second eradication regimen (ie, quadruple therapy) and a new 13C UBT 6 weeks after completion of therapy, for an extra cost of €87.93/patient. In the ITT analysis, taking into account the supplementary expenses due to eradication failure, the cost for each eradication achieved was almost comparable between CLA-ST and LEV500-ST (ie, €105.5 vs €106.9, respectively), but definitely less expensive (ie, cost-saving) with LEV250-ST (ie, €91.0 vs €105.5, LEV250-ST vs CLA-ST, respectively) (data not shown).
This randomised, controlled, multicentre study involving 375 patients demonstrates that in an area of high prevalence (ie, 15–20%) of CLA resistant H pylori strains, LEV-containing ST is more effective, equally safe, and cost-saving compared with CLA-containing ST.
Fluoroquinolones such as LEV have been used successfully in combination with PPI and AMO in the treatment of H pylori infection both as first-line25 26 and second-line26 27 therapy. In this study, LEV250-ST or LEV500-ST were significantly more effective than CLA-ST with eradication rates of 96% and 96.8%, respectively, compared with 80.8% using CLA-ST, in the ITT analysis. Other attempts have been made at evaluating LEV-containing ST. In particular, a recent study from Spain compared CLA and LEV in triple and sequential first-line therapies for H pylori eradication.28 The eradication rates of CLA-ST were 76.5% and 80.8%, whereas eradication rates of LEV containing ST were 82.5% and 85.2% in the ITT and PP analyses, respectively.28 No significant differences in compliance or adverse events were detected.28 The lower efficacy of LEV-ST compared with our study might be related to a higher prevalence of LEV resistant H pylori strains in Spain compared to our geographical area. In partial support of this hypothesis, a recent study by Cibrelus et al has shown an approximately 6% prevalence of LEV resistance in Spain29 which is almost twice as much the prevalence of LEV resistance in our area.
In our study, the efficacy of CLA-ST was somewhat lower than expected based on the results of original studies and subsequent metanalyses showing overall eradication rates of approximately 90%.12 13 15 30 This may, at least in part, be contributed to by the higher prevalence of CLA resistant (ie, approximately 20%) and dual (ie, CLA+MET) resistant (ie, approximately 6%) H pylori strains in our study population compared with that observed in those studies showing higher eradication rates with CLA-ST.12 30 In fact, Vaira et al showed a prevalence of CLA resistance and dual resistance of 12.5% and 4.3%, respectively.12 Similarly, in a recent study by Wu et al,30 the prevalence of CLA resistance was 6.6% and that of dual resistance 4.2%. In further support of this concept, in our study the efficacy of CLA-ST was 75% in the face of CLA resistance, approximately 90% in patients with CLA susceptible strains, and completely null in patients with dual resistance, which is consistent with data reported in a recent metanalysis by Zullo et al.13 Moreover, the efficacy achieved by CLA-ST in our study, that is 80.8% in the ITT analysis, is comparable to that described in more recent studies from Korea, Italy and Spain showing eradication rates of approximately 78%, 86% and 77%, respectively.21 22 28 Therefore, based on these observations, CLA-ST should no longer be recommended as first-line empirical treatment in areas of high prevalence of CLA-resistant H pylori strains.
The mechanism of action of ST is still unclear, but the initial dual therapy with AMO might reduce the bacterial load in the stomach and this, in turn, may improve the efficacy of the subsequently administered triple therapy.31 Also, it has been suggested that ST might be more effective than standard triple therapy because it involves the administration of four drugs (ie, PPI + three antimicrobials) with an additional antimicrobial (ie, tinidazole) compared with triple therapy.31 32 Therefore the question may arise as to whether an equivalent effect might be achieved by giving the four constituent of ST concomitantly in a 5-day period (ie, ‘concomitant’ therapy).33 This would make the four-drug approach easier for the patient to follow and would also be less expensive. Therefore, should the results of this study be corroborated by others, it may be hypothesised that the use of a LEV-containing 5-day concomitant therapy as the initial approach to H pylori infection, particularly in areas with high prevalence of CLA-resistant strains and low prevalence of LEV-resistant strains.
H pylori resistance to antimicrobials can be either primary (ie, existing before therapy) or secondary (ie, developing as the result of failed therapy). The incidence of secondary resistance has been dependent on the type of primary eradication therapy. Recently, we have shown that in our geographical area there is a dramatic increase in the prevalence of antimicrobial resistance in H pylori infected patients with previous unsuccessful eradication treatment.18 In particular, CLA resistance was about 45%, MET approximately 41% and LEV 14%. Also, eradication failure caused the emergence of H pylori strains with multiple resistance.18 Therefore, one of the main issues today in H pylori therapy is to decrease the number of eradication failures as much as possible.34 We postulate that the significantly lower rate of treatment failures observed following LEV-containing STs compared with CLA-ST might contribute to prevention of the emergence of bacterial secondary resistance.
Another limitation to 100% efficiency of eradication regimens is compliance to treatment mainly due to drug-related adverse events. In our study, patients were instructed not to suspend therapy arbitrarily and to contact us when adverse events occurred. All treatments were well tolerated with an incidence of adverse events comparable between three groups. Major adverse events were diarrhoea and epigastric pain mainly during the first period of treatment and bad/bitter taste during the second period of treatment. Only in one case was diarrhoea so severe as to cause discontinuation of therapy. Also, we only had two patients who developed glossitis during the second period of treatment and this was in the CLA-ST group. Fluoroquinolones have been reported to cause muscle or tendon pain. In our study only one patient in the LEV500-ST group reported muscular pain to the limbs but this was not so severe to cause discontinuation of therapy.
Cost is an important consideration for any therapy. However, cost cannot be considered separately from effectiveness. LEV-containing STs were more expensive than CLA-ST, but, when considering the difference in effectiveness, the cost per eradication with LEV250-ST was almost €15 (ie, US$25) lower than the cost per eradication with CLA-ST. This was due to the costs of additional office visit, new eradication regimen and new 13C UBT following treatment failure. The difference in cost-effectiveness between the two regimens might even be higher if we consider indirect costs of eradication failure such as unfavourable clinical outcomes, including gastric atrophy, peptic ulcer or gastric cancer, lost time from work and quality of life, variables which are difficult to measure and, thus, often ignored. However, this information cannot be completely generalised to other countries which might have different costs per each of the item evaluated in our study.
In conclusion, this randomised trial demonstrates that in an area of high (ie, >15%) prevalence of CLA-resistant H pylori strains, LEV-containing ST is well tolerated and significantly superior to CLA-ST. Finally, LEV-containing ST is cost saving, decreasing the need for supplementary office visit, new eradication regimen and re-testing. We postulate that, because antimicrobial-resistant strains are becoming increasingly prevalent, LEV-containing sequential regimens might be used as first-line therapy in geographical areas with higher than 15% prevalence of H pylori CLA-resistant strains in order to improve the efficacy of H pylori eradication therapy and prevent the emergence of secondary resistance. However, it must be kept in mind that the main disadvantage with use of LEV-containing therapies is the rapid emergence of resistance to this antimicrobial and that in some part of Europe high prevalence of LEV-resistant strains has been described.35 Therefore, before recommending LEV-containing therapies as the first-line empirical choice, one must be confident that the local prevalence of LEV-resistant strains is low.
Competing interests None.
Ethics approval This study was conducted with the approval of the Ethics Committee of the Second University of Naples, Italy.
Provenance and peer review Not commissioned; externally peer reviewed.
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.