Randomized controlled study of rabeprazole, levofloxacin and rifabutin triple therapy vs. quadruple therapy as second-line treatment for Helicobacter pylori infection
Dr B. C. Y. Wong, Department of Medicine, University of Hong Kong, Queen Mary Hospital, Hong Kong, China.
Aim : To test the efficacy of rabeprazole, levofloxacin and rifabutin triple therapy vs. quadruple therapy for the second-line treatment of Helicobacter pylori infection.
Methods : One hundred and nine patients who had failed previous H. pylori eradication were randomized to receive: (i) rabeprazole, 20 mg b.d., rifabutin, 300 mg once daily, and levofloxacin, 500 mg once daily, for 7 days (triple therapy); or (ii) rabeprazole, 20 mg b.d., metronidazole, 400 mg t.d.s., bismuth subcitrate, 120 mg q.d.s., and tetracycline, 500 mg q.d.s., for 7 days (quadruple therapy). Endoscopy and culture were performed before treatment.
Results : The clarithromycin (79% vs. 21%, P < 0.001) and metronidazole (89% vs. 40%, P < 0.001) resistance rates were significantly higher in patients with previous exposure than in those with no previous exposure. The intention-to-treat and per protocol eradication rates were 91%/91% for the triple therapy group and 91%/92% for the quadruple therapy group. For patients with double resistance to metronidazole and clarithromycin, the eradication rates were 85% (17/20) in the triple therapy group and 87% (13/15) in the quadruple therapy group. Compliance was greater than 95% for both regimens.
Conclusion : Rabeprazole, levofloxacin and rifabutin-based triple therapy and quadruple therapy were equally effective as second-line treatments for H. pylori infection.
Helicobacter pylori infection plays an important role in the pathogenesis of peptic ulcer disease, maltoma and adenocarcinoma of the stomach.1 The eradication of H. pylori significantly reduces the relapse of peptic ulcers.2–5 Bacterial resistance to metronidazole or clarithromycin is an important factor leading to treatment failure. Primary resistance to metronidazole6 and clarithromycin7 is common in our region and significantly affects the effectiveness of standard eradication therapy.8, 9 Several consensus meeting reports, including the Asia-Pacific Consensus Conference on the Management of H. pylori Infection and the Maastricht 2-2000 Consensus Report, have recommended the use of quadruple therapy (proton pump inhibitor, bismuth, tetracycline, metronidazole) for 1 week as second-line therapy for H. pylori infection.1, 10 However, the dosing schedule of quadruple therapy is complicated and the eradication rates of re-treatment are in the range 57–95%.11–18 Most of these studies were limited by the small number of patients,11–15 provided no antibiotic sensitivity data11–13, 15, 17, 18 and were non-randomized, single-arm studies.11–17 Rabeprazole, a newly developed benzimidazole proton pump inhibitor, is a more potent inhibitor of H+,K+-ATPase and acid secretion than omeprazole, and is a more rapid inhibitor of proton pumps than omeprazole, lansoprazole and pantoprazole.19, 20In vitro studies have demonstrated that it has a more potent antibacterial activity against H. pylori when compared with either omeprazole or lansoprazole.21, 22 Recently, rifabutin, a spiropiperidyl derivative of rifamycin, has been shown to exhibit high in vitro activity against H. pylori.23–25 Furthermore, rifabutin is chemically stable over a wide pH range and its antibacterial activity is not affected by the acidic environment of the stomach.26 Preliminary clinical trials have suggested that rifabutin may be a promising second-line agent for the treatment of H. pylori infection, with an eradication rate of around 70%.27, 28 Similarly, levofloxacin, a laevorotatory isomer of ofloxacin, has shown an excellent activity against a variety of Gram-positive and Gram-negative organisms which are resistant to the established agents.29 The bio-availability of levofloxacin approaches 100% and is little affected by the co-ingestion of food. A regimen using levofloxacin, 500 mg once daily, plus rabeprazole, 20 mg once daily, and either amoxicillin, 1000 mg b.d., or tinidazole, 500 mg b.d., as first-line therapy for 1 week has been shown to achieve an eradication rate of 90–92%.30 Thus, we designed a new triple therapy combination of rabeprazole, rifabutin and levofloxacin for use as second-line treatment of H. pylori infection.
The aims of this randomized controlled study were: (i) to evaluate the efficacy of 1-week rabeprazole, levofloxacin and rifabutin triple therapy vs. rabeprazole-based quadruple therapy for the treatment of H. pylori infection after failure of one or more courses of H. pylori eradication; (ii) to determine the side-effect profile; and (iii) to assess the influence of antibiotic resistance on the treatment efficacy.
Consecutive patients presenting to the Queen Mary Hospital, Hong Kong, and satisfying the following criteria were invited to participate in this study: age greater than 18 years; failure of at least one course of H. pylori eradication therapy; agreement to undergo repeat endoscopy and biopsies for H. pylori culture. Informed written consent was obtained from all patients who participated in the trial. Exclusion criteria included patients who had been taking aspirin, other non-steroidal anti-inflammatory drugs (NSAIDs), antibiotics, H2-receptor blockers, bismuth or proton pump inhibitors in the preceding 4 weeks, known drug allergy to the study drugs, severe cardio-pulmonary, liver or renal diseases, pregnancy or lactation and previous gastric surgery. This study was approved by the ethics committee of the University of Hong Kong.
Diagnosis of H. pylori infection
During endoscopy, three antral biopsies and one corpus biopsy were taken. One antral biopsy was sent for H. pylori culture, one antral biopsy was used for rapid urease test and the rest were sent for histological examination of H. pylori status by haematoxylin and eosin stain and Giemsa stain if necessary. Specimens were read by experienced pathologists who were blind to all clinical information. The definition of H. pylori infection in this study required at least two positives out of the three tests (CLO test, histology and culture).
Culture and antibiotic resistance
Antral biopsies were collected during upper endoscopy and transported to the laboratory immediately. H. pylori was cultured from gastric biopsy specimens on selective media (Columbia agar with 7% horse blood and H. pylori-selective supplement; Oxoid, Basingstoke, UK) under micro-aerophilic conditions produced by a gas-generating system (CampyGen; Oxoid, Basingstoke, UK) for 3–6 days. H. pylori was confirmed by Gram staining and by positive urease, oxidase and catalase tests. Metronidazole, clarithromycin and amoxicillin susceptibilities were assessed using the E-test (AB Biodisk, Sweden). Metronidazole resistance was defined as a minimal inhibitory concentration of > 8 µg/mL. Clarithromycin and amoxicillin resistance was defined as a minimal inhibitory concentration of > 2 µg/mL.
Treatment regimen and follow-up
The patients were randomized to receive one of the following treatments: (i) triple therapy consisting of rabeprazole, 20 mg b.d., rifabutin, 300 mg once daily, and levofloxacin, 500 mg once daily, for 7 days; or (ii) quadruple therapy consisting of rabeprazole, 20 mg b.d., bismuth subcitrate, 120 mg q.d.s., metronidazole, 400 mg t.d.s., and tetracycline, 500 mg q.d.s., for 7 days. Randomization was performed by drawing a sealed envelope that contained a pre-assigned randomized treatment number generated by computer on entry to the study. Study drugs with dosing instructions were packed into sealed envelopes with an assigned treatment number. The patients were given a diary to record the side-effects and symptoms during therapy. Side-effects were also documented by a direct questioning approach by an investigator who was blind to the assigned treatment of the patient 1 week after treatment. 13C-Urea breath test was performed 6 weeks after drug treatment. The technician who performed the breath test was blind to the assigned treatment of the patient. Briefly, the patient fasted for 4 h before the test. No test meal was given and a pre-dose breath sample was obtained. 13C-Urea powder (75 mg) dissolved in 50 mL of water was given orally. The second breath sample was collected at 30 min. The cut-off value used was 5‰. All patients were kept in the sitting position over the whole study period. Collected samples were analysed by a purpose-built isotope ratio mass spectrometer. This protocol has been validated in our centre with a sensitivity and specificity of 96.5% and 97.7%, respectively.31 Those who failed H. pylori eradication were offered a rescue treatment by crossing over to the other treatment arm after unblinding of the treatment code.
In view of the high prevalence of metronidazole and clarithromycin resistance of H. pylori in our region, an expected efficacy of 65% successful eradication was adopted. To detect a 25% difference in efficacy of the tested regimen with a power of 80% and an α error of 5%, at least 51 patients in each arm are required. The statistics used included the chi-squared test, Fisher's exact test, Student's t-test and Mann–Whitney U-test for data with a skewed distribution. A P value of 0.05 or less was considered to be statistically significant. All P values were two-sided. The intention-to-treat analysis included all patients who had taken at least one tablet of the drug. In the per protocol analysis, patients with poor drug compliance (< 70% intake of any study drug) and defaulters were excluded.
Demographic data (Table 1)
Table 1. Demographic data of the patients
|No. of patients (intention-to-treat)||56||53|| |
|Age (years)||45 ± 11||46 ± 12||0.64|
|Previous failed treatment (mean)|| 1.4|| 1.3||0.55|
|Smoking (%)|| 5||15||0.12|
|Alcohol (%)||11|| 9||1.0|
| Non-ulcer dyspepsia||43||37|| |
| Gastroduodenal erosions|| 3|| 1|| |
| Duodenal ulcer|| 7||14|| |
| Gastric ulcer|| 3|| 1||0.54|
|No. of patients excluded|
| Defaulted follow-up|| 0|| 1|| |
| Non-compliance|| 2|| 0|| |
|No. of patients (per protocol)||56||52|| |
One hundred and nine patients were recruited (mean age, 45 years; range, 19–74 years). There were 43 males (mean age, 48 years) and 66 females (mean age, 44 years). A history of smoking and alcohol use was present in 11 (10%) and 11 (10%) patients, respectively. Seventy-five patients (69%) had failed one course of H. pylori eradication before and 34 patients (31%) had failed two or more courses before. The majority of patients had functional dyspepsia (73%). Twenty-one patients (19%) had duodenal ulcer, four (4%) had gastric ulcer and four (4%) had gastroduodenal erosions. One patient was lost to follow-up and two patients took less than 70% of the study drugs and were excluded from the per protocol analysis.
Table 2. Helicobacter pylori eradication rates and antibiotic resistance in relation to previous treatment regimens
|(A) PPI/Rbc + Cla + Amo||9/28 (32)||21/28 (75)||8/28 (29)||17/17 (100)||16/17 (94)||33/34 (97)|
| (n = 34, positive culture = 28)|
|(B) PPI/Rbc + Amo + Met||16/16 (100)||3/16 (19)||3/16 (19)||12/14 (86)||8/9 (89)||20/23 (87)|
| (n = 23, positive culture = 16)|
|(C) PPI/Rbc + Cla + Met||8/9 (89)||7/9 (78)||7/9 (78)||3/4 (75)||7/7 (100)||10/11 (91)|
| (n = 11, positive culture = 9)|
|(D) Others†||4/6 (67)||2/6 (33)||2/6 (33)||2/2 (100)||4/5 (80)||6/7 (86)|
| (n = 7, positive culture = 6)|
|(E) Two courses or more‡||17/23 (74)||21/23 (91)||15/23 (65)||17/19 (89)||13/15 (87)*||30/34 (88)|
| (n = 34, positive culture = 23)|
|Total||54/82 (66)||54/82 (66)||35/82 (43)||51/56 (91)||48/53 (91)||99/109 (91)|
| (n = 109, positive culture = 82)|
Table 3. Intention-to-treat Helicobacter pylori eradication rates in relation to treatment group and antibiotic susceptibility
|Eradication rate (%)|
|MetS||2/2 (100)||7/8 (88)||9/10 (90)||6/7 (86)||11/11 (100)||17/18 (94)|
|MetR||12/13 (92)||17/20 (85)||29/33 (88)||5/6 (83)||13/15 (87)*||18/21 (86)|
|Total||14/15 (93)||24/28 (86)||38/43 (88)||11/13 (85)||24/26 (92)||35/39 (90)|
Positive H. pylori culture was available in 90 patients (83%) and subculture for antibiotic sensitivity in 82 patients (75%). Resistance to metronidazole was found in 54 patients (66%) and resistance to clarithromycin in 54 patients (66%). No strains developed resistance to amoxicillin. Double resistance to metronidazole and clarithromycin was found in 35 patients (43%). For patients who had received one course of standard triple therapy before, the metronidazole resistance rate ranged from 89% to 100% after failure of metronidazole-containing regimens (regimens B and C) and the clarithromycin resistance rate ranged from 75% to 78% after failure of clarithromycin-containing regimens (regimens A and C). After one course of standard triple therapy, double resistance to metronidazole and clarithromycin was 19% for the metronidazole-containing regimen and 29% for the clarithromycin-containing regimen. However, the double resistance rate increased to 78% after a regimen containing both metronidazole and clarithromycin. After the failure of two or more courses of eradication treatment, the metronidazole resistance rate was 74% and the clarithromycin resistance rate was 91% (metronidazole usage was 76% and clarithromycin usage was 100% in the previous regimens) (Table 2). The overall clarithromycin resistance rate was significantly higher in patients who had used a clarithromycin-containing regimen when compared with those who had not (79% vs. 21%, P < 0.001). Similarly, the overall metronidazole resistance rate was significantly higher in patients who had used a metronidazole-containing regimen when compared with those who had not (89% vs. 40%, P < 0.001).
According to the intention-to-treat analysis, at week 6, H. pylori eradication was achieved in 51 patients in the triple therapy group and in 48 patients in the quadruple therapy group. The intention-to-treat and per protocol eradication rates were 91% and 91% in the triple therapy arm and 91% and 92% in the quadruple therapy group (P = 1.0). For patients who had failed one course of H. pylori eradication therapy, the overall intention-to-treat eradication rate was 92% in the triple therapy group and 92% in the quadruple therapy group. For patients who had failed two or more courses of H. pylori eradication therapy, the intention-to-treat eradication rate was 89% in the triple therapy group and 87% in the quadruple therapy group. Table 3 shows the intention-to-treat analysis of the two regimens with respect to metronidazole and clarithromycin susceptibility. Both the triple and quadruple therapy regimens achieved a high H. pylori eradication rate despite the presence of metronidazole and clarithromycin resistance [antibiotic sensitivity data available in 82/109 (75%) of patients]. For patients with double resistance to metronidazole and clarithromycin, the eradication rates were 85% (17/20) in the triple therapy group and 87% (13/15) in the quadruple therapy group.
Side-effects and compliance (Table 4)
Table 4. Side-effects in each treatment group
|Nausea/vomiting (%)||3† (5)||7.0||15 (28)||6.3|
|Loose stool/diarrhoea (%)||3* (5)||7.0||10 (19)||6.9|
|Dizziness (%)||7 (13)||6.1||8 (15)||6.4|
|Headache (%)||5 (9)||6.2||2 (4)||4.5|
|Taste disturbance (%)||3 (5)||7.0||4 (8)||7.0|
|Malaise (%)||1 (2)||7.0||4 (8)||6.3|
|Others (%)||4 (7)||7.0||6 (11)||7.0|
|Total patients (%)||19 (34)*|| ||31 (58)|| |
Drug compliance was excellent. Ninety-six per cent of patients in the triple therapy group and 98% of patients (one patient defaulted follow-up) in the quadruple therapy group completed the study medications. One female patient in the triple therapy group developed drug-related neutropenia (white blood cell count, 0.8 × 109/L) and thrombocytopenia (platelet count, 68 × 109/L) 1 week after treatment. Bone marrow trephine biopsy showed a normocellular marrow and suggested a peripheral cause of cytopenia. The patient had an uneventful recovery after conservative treatment and reverse isolation for 7 days. The most common side-effects were dizziness and headache in the triple therapy group and nausea/vomiting and loose stool/diarrhoea in the quadruple therapy group. The number of patients having side-effects (34% vs. 58%, P = 0.02) and the mean duration (2.5 days vs. 3.9 days, P = 0.04) of all side-effects were significantly lower in the triple therapy group than in the quadruple therapy group.
Factors affecting H. pylori eradication
The mean age of the patients who failed H. pylori eradication was 48.1 years vs. 45 years in patients with successful eradication (P = N.S.). Sex, smoking and drinking had no effect on the outcome of eradication. The presence of metronidazole and clarithromycin resistance did not affect the success of H. pylori eradication by the two regimens.
Six (three in each group) of the 10 patients who failed H. pylori eradication agreed to be re-treated by crossing over (triple to quadruple therapy or vice versa) to the other treatment arm. Post-treatment 13C-urea breath test at week 6 was negative in all six patients.
We have reported a randomized controlled study of rabeprazole, levofloxacin and rifabutin triple therapy vs. rabeprazole-based quadruple therapy for the treatment of H. pylori infection after failure of one or more courses of previous H. pylori eradication therapy. The intention-to-treat and per protocol H. pylori eradication rates were 91%/91% for the triple therapy group and 91%/92% for the quadruple therapy group. The clarithromycin (79% vs. 21%, P < 0.001) and metronidazole (89% vs. 40%, P < 0.001) resistance rates were significantly higher in patients who had taken these antibiotics in previous eradication regimens. However, the eradication rates of both regimens were satisfactory despite the presence of single or dual antibiotic resistance to metronidazole and clarithromycin.
Previous non-randomized studies have shown that rifabutin, 300 mg daily, in combination with proton pump inhibitor and amoxicillin for 1 week achieves a satisfactory intention-to-treat eradication rate of around 70% after failure of standard triple therapy.27, 28 In a randomized controlled study comparing 10 days of rifabutin (300 mg daily or 150 mg daily) plus pantoprazole and amoxicillin with pantoprazole-based quadruple therapy (bismuth, tetracycline and metronidazole) for 10 days, the intention-to-treat eradication rates were 67% in the rifabutin 150 mg group and quadruple therapy group, and 87% in the rifabutin 300 mg group.32 Our data suggest that, by the addition of levofloxacin, an excellent eradication rate of 91% can be achieved. This regimen was effective even for patients who had failed two or more courses of eradication therapy (eradication rate of 89%) and in the presence of dual metronidazole and clarithromycin resistance (eradication rate of 85%).
Quadruple therapy (proton pump inhibitor, bismuth, tetracycline, metronidazole) for 1 week has been recommended as optimal second-line therapy at major consensus conferences.1, 10 A pooled analysis of 24 trials of quadruple regimens (proton pump inhibitor or H2 blocker, bismuth, tetracycline, metronidazole) as second-line therapy reported a pooled eradication rate of 74%.33 The dosage of metronidazole used is important in determining the success rate of the quadruple regimen.34 We have shown that the use of a high dose of metronidazole, frequent dosing of bismuth and a proton pump inhibitor with more potent anti-H. pylori activity, together with high patient compliance, achieves an excellent eradication rate of 91% with a 7-day quadruple regimen. In the randomized controlled trial of rifabutin-based triple therapy vs. quadruple therapy for 10 days, described above, the dosage of metronidazole was only 750 mg daily and bismuth was given twice daily rather than four times daily, resulting in an eradication rate of 67% only.32 Furthermore, we found that our triple and quadruple regimens were complementary. Those who failed the triple therapy regimen could be re-treated with the quadruple therapy regimen and vice versa. Although the total number of patients who received the ‘rescue treatment’ was small, our study suggests that this is a feasible approach for patients with ‘uneradicable’ infection.34
The clarithromycin resistance rate was significantly higher in patients who had used a clarithromycin-containing regimen before, suggesting that the re-administration of clarithromycin after failure of a clarithromycin-containing regimen is not advisable.35 Post-treatment dual resistance to metronidazole and clarithromycin ranged from 19% to 78% and was highest in patients who had received a metronidazole and clarithromycin combination as first-line therapy. As post-treatment antibiotic resistance to metronidazole and clarithromycin is common, it has been suggested that the antibiotics of first choice in first-line therapy should not combine clarithromycin and metronidazole.34–36
The proportion of patients with side-effects and the mean duration of side-effects were lower in the triple therapy group when compared with the quadruple therapy group. In general, both regimens were well tolerated. However, one patient in the triple therapy group developed drug-related neutropenia and thrombocytopenia 1 week after treatment, which are side-effects of rifabutin.37, 38 Although this is a rare event, physicians should be aware of these potentially serious side-effects when considering the rifabutin-based regimen. As rifabutin is very expensive (cost of the triple therapy regimen was US$128 vs. US$ 25 for the quadruple therapy regimen in Hong Kong) and is a member of the group of anti-mycobacterial drugs against multi-resistant Mycobacterium tuberculosis, this drug should be reserved only for resistant cases of H. pylori infection or even as third-line therapy.
Our study has two limitations. One caveat is that the primary treatment regimen in our study was not controlled, but reflected the real life situation in clinical practice. Secondly, the two treatments were not totally identical in terms of the number of tablets taken and the dosing interval. However, the satisfactory drug compliance observed in the present trial suggests that this is not a significant problem.
In conclusion, 1-week rabeprazole, levofloxacin and rifabutin-based triple therapy and rabeprazole-based quadruple therapy were equally effective for the treatment of H. pylori infection after the failure of one or more courses of H. pylori eradication. Both therapies were effective despite the presence of double resistance to metronidazole and clarithromycin.
The authors thank Nurse Specialist Ms M. Chong, Endoscopy Nurses Ms K. W. Wong, Vera S. Y. Tang, Diana K. K. Chang and W. P. Yung and research assistants T. S. M. Tong and V. Y. K. Ho for data management and assistance. This study was supported by an unrestricted grant from Eisai Hong Kong Ltd, the Simon K. Y. Lee Gastroenterology Research Fund and the Peptic Ulcer Research Fund, University of Hong Kong, Hong Kong.