- Email: [email protected]
Levofloxacin- versus metronidazole-based rescue therapy for H. pylori infection in Japan
Digestive and Liver Disease 37 (2005) 821–825
Alimentary Tract
Levofloxacin- versus metronidazole-based rescue therapy for H. pylori infection in Japan Y. Matsumoto a , I. Miki a , N. Aoyama b,∗ , D. Shirasaka a , Y. Watanabe a , Y. Morita a , M. Toyoda a , T. Mitani a , H. Miyachi a , T. Tamura a , M. Kasuga a a
b
Division of Diabetes, Digestive and Kidney Diseases, Department of Clinical Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan Department of Endoscopy, Kobe University Graduate School of Medicine, 7-5-1, Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan Received 26 April 2005; accepted 14 June 2005 Available online 22 July 2005
Abstract Background. The ideal second-line treatment regimens for Helicobacter pylori infection may differ between the areas, countries and races. Aim. The aim was to confirm which was the better regimen for second-line therapy after treatment failure with a standard triple therapy in Japan, a high dosage of levofloxacin- or metronidazole-based therapy. Patients. Sixty outpatients with persistent H. pylori infection after a standard triple therapy were enrolled in this prospective, open-label and randomised trial. Methods. The subjects were randomly administered levofloxacin (300 mg b.d.)- or metronidazole (500 mg b.d.)-based therapy with lansoprazole (30 mg b.d.) and amoxicillin (1000 mg b.d.) for 7 days, and the cure rates and side effects were analysed. Antimicrobial susceptibility was also examined before second-line therapy using the E-test. Results. Good compliance was obtained without severe side effects in both the groups except for two patients. The cure rates, expressed as intention-to-treat and per-protocol analyses, respectively, were 70.0 and 72.4% in the levofloxacin group, and 96.7 and 100% in the metronidazole group. Each regimen often overcame even clarithromycin-resistant strains. Conclusion. Metronidazole-based triple therapy is recommended as second-line therapy in Japan, and levofloxacin-based therapy can be an alternative treatment option. © 2005 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. Keywords: H. pylori; Levofloxacin; Metronidazole; Second-line
1. Introduction Proton pump inhibitor (PPI)-based triple therapy, containing two antibiotics among amoxicillin (AMOX), clarithromycin (CAM) and metronidazole (MNZ), is widely considered the golden standard for first-line therapy of H. pylori infection [1–3]. However, the therapeutic efficacy varies widely, ranging 25–95% [4–8]. Recent years have witnessed a decrease in its cure rate due to the progres∗ Corresponding author. Tel.: +81 78 382 5863/6570; fax: +81 78 382 6579. E-mail address: [email protected] (N. Aoyama).
sion of antimicrobial resistance and currently, individuals with refractory H. pylori infection cannot be ignored [9–13]. Reportedly, poor compliance and antimicrobial resistance, in particular CAM or MNZ resistance, are considered some of the most important factors of treatment failure [9,11,14]. It is, therefore, an urgent issue to develop an appropriate eradication strategy. The best rescue treatment regimens should differ among countries and races because of differences in antimicrobial resistance patterns. In Japan, PPI with AMOX and CAM is recommended by the Japanese Helicobacter Society as first-line therapy against H. pylori infection [15], and the treatment efficacy of this regimen mainly depends on CAM
1590-8658/$30 © 2005 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dld.2005.06.002
822
Y. Matsumoto et al. / Digestive and Liver Disease 37 (2005) 821–825
susceptibility [16,17]. In case of treatment failure, no fixed regimen has been established for second-line therapy in Japan. The ideal rescue regimen should consist of short-term and simple drug combinations using susceptible agents with good compliance and little toxicity. Of several rescue regimens proposed in various countries, the combination of PPI, AMOX and MNZ is often administered as second-line treatment in Japan, where MNZ resistance is relatively uncommon compared to worldwide tendency [18–22]. The cure rate of the 7-day PPI–AMOX–MNZ regimen as second-line therapy is around 90% in Japan [19,20,22]. This regimen has the advantage of compliance because of its simple and short-term treatment. However, MNZ resistance may also be widespread in Japan due to its increased consumption, as with CAM resistance. Furthermore, dual resistance to CAM and MNZ is often detected [21,23]. Therefore, it is important to find a new effective antibiotic with low toxicity instead of CAM or MNZ. Of several candidates a new fluoroquinolone, levofloxacin (LVFX), is considered an attractive agent for second-line treatment due to its excellent activity against H. pylori [23,24]. Our previous study revealed that LVFX has strong activity even against CAM-resistant H. pylori strains in vitro, and the efficacy and safety of LVFX 400 mg-based therapy were confirmed as second-line therapy, showing a successful eradication rate of 70% [24]. According to an Italian trial, LVFX 500 mg-based therapy showed a high re-eradication rate of 94% [25]. Focusing on the difference in LVFX dosage, we hypothesised that a high dosage of LVFX-based therapy could induce increased efficacy, and that this regimen might compare with MNZ-based regimen in efficacy and toxicity in the Japanese population. This study was carried out to confirm which was the ideal second-line regimen for refractory H. pylori infection to a standard triple therapy in Japan, a high dosage of LVFX-based therapy or MNZ-based therapy.
2. Patients and methods This study was carried out in a prospective, open-label, randomised, parallel group fashion. The study subjects were 60 consecutive outpatients with persistent H. pylori infection after a standard triple therapy (gastric ulcer, n = 26; duodenal ulcer, n = 14; gastroduodenal ulcer, n = 4; gastritis, n = 16; 36 males and 24 females; mean age, 53.1 ± 13.7). These patients were enrolled according to the eligibility criteria (age greater than 20 years and younger than 70 years) and the exclusion criteria (patients who had been taking aspirin, other non-steroidal anti-inflammatory drugs, known drug allergy to the study drugs, gastric cancer, severe concomitant disease and previous gastric surgery). All the patients underwent endoscopy before first-line therapy to confirm gastroduodenal disease and received 7-day PPI–AMOX–CAM therapy as first-line therapy. H. pylori status was confirmed by a 13 C-urea breath test before the trial. The cut-off value was
Table 1 Clinical characteristics of LVFX and MNZ groups LVFX group (n = 30)
MNZ group (n = 30)
Age (years) Male/female
50.8 ± 13.5 17/13
52.0 ± 13.0 19/11
Disease GU DU GDU Gastritis
15 6 2 7
11 8 2 9
First-line regimen LPZ–AMOX–CAM OPZ–AMOX–CAM RPZ–AMOX–CAM
27 1 2
28 1 1
Smoking/non-smoking Drinking/non-drinking AMOX-S/-R/unknown CAM-S/-R/unknown LVFX-S/-R/unknown MNZ-S/-R/unknown
8/22 13/17 17/0/13 5/12/13 15/2/13 15/2/13
14/16 16/14 18/0/12 9/9/12 15/3/12 17/1/12
GU, gastric ulcer; DU, duodenal ulcer; GDU, gastroduodenal ulcer; LPZ, lansoprazole 30 mg b.d.; AMOX, amoxicillin 750 mg b.d.; CAM, clarithromycin 200 mg or 400 mg b.d.; RPZ, rabeprazole 20 mg b.d.; OPZ, omeprazole 20 mg b.d.; -S, suspeptible; -R, resistant.
set at 2.5‰ 20 min after taking 100 mg 13 C-urea [26]. Follow-up endoscopy was performed for a biopsy to examine the antimicrobial susceptibility of H. pylori before secondline treatment when agreement was obtained. The antimicrobial susceptibility testing was assessed using the E-test (AB Biodisk, Solna, Sweden) as previously described [17]. AMOX, CAM, LVFX and MNZ resistance were defined as the minimum inhibitory concentration of >8, >1, >1 and >8 g/ml, respectively. The result of the susceptibility testing was not available before the evaluation of the secondline treatment outcome. After assessment for refractory H. pylori infection to standard PPI-based triple therapy, these patients were randomly assigned by computer-generated random numbers to receive one of the following treatments: (a) 300 mg LVFX b.d., 20 mg lansoprazole (LPZ) b.d. and 1000 mg AMOX b.d. for 7 days (LVFX group; n = 30), or (b) 500 mg MNZ b.d., 20 mg lansoprazole b.d. and 1000 mg AMOX b.d. for 7 days (MNZ group; n = 30). The clinical characteristics of each group are shown in Table 1. Compliance and side effects were documented by a detailed questionnaire sheet during treatment. The patients were advised to fill in the sheet every day, which consisted of several columns expressing the drug intake time and the grade (mild–moderate–severe) of possible side effects (nausea, vomiting, abdominal discomfort, soft stool, diarrhoea, constipation, loss of appetite, taste disturbance, headache, vertigo, general fatigue, skin rash and others). A 13 C-urea breath test was performed 8 weeks after the completion of second-line therapy. Treatment efficacy was expressed using both intentionto-treat (ITT) and per-protocol (PP) analyses. The 95%
Y. Matsumoto et al. / Digestive and Liver Disease 37 (2005) 821–825 Table 2 Side effects of LVFX and MNZ groups
823
Table 3 PP cure rate according to antimicrobial resistance pattern
LVFX group
MNZ group
Diarrhoea/soft stool Nausea General fatigue Abdominal discomfort Skin rash
3 1 1 1 1
6 3 1 1 0
Total episodes Total patients (%)
7 6 (20.0)
11 8 (26.7)
LVFX, levofloxacin; MNZ, metronidazole.
confidence intervals (CI) were also calculated. The Fisher’s exact test or χ2 -test were used to compare the differences in clinical characteristics, side effects and eradication rates of each group. A p-value less than 0.05 was considered significant. This study was approved by the Interstitial Review Board of Kobe University Hospital. Written informed consent was obtained from all subjects before their enrollment in this study.
LVFX group CAM-S, LVFX-S CAM-S, LVFX-R CAM-R, LVFX-S CAM-R, LVFX-R Unknown Total
3/4 (75.0%) – 6/10 (60.0%) 1/2 (50.0%) 11/13 (84.6%) 21/29 (72.4%)
MNZ group CAM-S, MNZ-S CAM-S, MNZ-R CAM-R, MNZ-S CAM-R, MNZ-R Unknown Total
8/8 (100%) 1/1 (100%) 8/8 (100%) – 12/12 (100%) 29/29 (100%)
LVFX, levofloxacin; MNZ, metronidazole; CAM, clarithromycin; -S, susceptible; -R, resistant.
ture in the LVFX group, and all the patients without culture in the MNZ group showed successful eradication.
4. Discussion 3. Results Sixty patients were enrolled in this trial and were divided into two groups. The two groups were well matched with regard to demographic and clinical characteristics (Table 1). All patients completed the therapeutic regimens except for two: one patient in the LVFX group dropped out because of mild skin rash, and another in the MNZ group because of severe diarrhoea. Except for these two, the side effects were very mild in both the groups, and the patients maintained good compliance. The overall prevalence of side effects was similar and no specific adverse effects were found in either regimen (Table 2). The cure rate, expressed as ITT and PP analyses, respectively, was 70.0% (95% CI = 45–95) and 72.4% (95% CI = 56–89) in the LVFX group and 96.7% (95% CI = 90–100) and 100% in the MNZ group. The treatment outcome showed a significantly high cure rate in the MNZ group compared to that in the LVFX group (p = 0.006 in ITT analysis and p = 0.002 in PP analysis). H. pylori susceptibility testing was performed in 38 patients before second-line treatment and the results of the E-test were successfully obtained in 35 patients, including the two patients who dropped out. The antimicrobial resistance pattern of each group was shown in Table 1. The overall CAM, LVFX and MNZ resistance were found in 21 (60.0%), five (14.2%) and three (8.6%) out of 35 examined, respectively. AMOX resistance was not found. Table 3 shows the PP cure rate according to each antimicrobial resistance pattern. MNZ-based therapy could eradicate all the strains, even MNZ- and CAM-resistant strains. On the other hand, LVFX-based therapy could eradicate seven out of 12 (58.3%) CAM-resistant strains. Eleven out of 13 patients without cul-
This trial was conducted in order to establish a simple and short-term regimen with effective and non-toxic agents for an ideal second-line treatment for H. pylori infection in Japan. Many investigators have identified attractive agents, such as rifabutin and furazolidone; however, these have limited use in consideration of toxicity, specific utility and cost benefit [27–29]. Of several other candidates, we focused on LVFX. PPI with LVFX and AMOX is a promising regimen for the salvage therapy [24,25,30–32]. These studies support that LVFX-based therapy is an attractive treatment option; however, only few data are available concerning LVFX susceptibility in the clinical trials [24,33]. Our previous trial using LVFX 400 mg-based regimen did not show entirely satisfactory re-eradication rate [24]. Therefore, this trial was conducted using a high dosage of LVFX (600 mg) in order to approach the high cure rate of LVFX 500 mg-based regimens reported by Italian groups [25]. In contrast to our expectation, however, the cure rate was not higher than that of our previous trial with LVFX 400 mg (72.4% versus 69.7% according to PP analysis) [24], or the Italian groups’ report (72.4% versus 94% according to PP analysis) [25] and this regimen was not superior to MNZ-based triple therapy (72.4% versus 100% according to PP analysis), although it may not be appropriate to compare the results of different trials. In addition, MNZ has the advantage of low cost compared to LVFX. One tablet of MNZ and LVFX costs 42 and 212 Japanese yen, respectively. LVFX resistance was not rare in this trial(14.2%). LVFXbased triple therapy could often eradicate even CAMresistant strains. However, this regimen could not sometimes eradicate even LVFX-susceptible strains (cure rate of only 60.0% according to PP analysis), although LVFX showed excellent activity in vitro [23,24]. It cannot be denied that the cure rate of LVFX-based therapy in this trial may be estimated
824
Y. Matsumoto et al. / Digestive and Liver Disease 37 (2005) 821–825
lower than the actual efficacy because of the limited number of patients. The results obtained can be discussed according to the following factors: first, LVFX susceptibility judged by the E-test may not always predict the treatment outcome. As for MNZ, there may be a discrepancy between the E-test and the agar dilution method [34,35]. As for CAM, the results of susceptibility obtained by the culture method may be different from those obtained by the PCR method [36]. In this study, the PP cure rate of LVFX-susceptible strains was only 64.3%, which was lower than that of all the patients treated with LVFX (72.4%). The same content as found in MNZ and CAM may also apply to LVFX. The LVFX-resistance mechanism is screened by DNA gyrase mutations, which can be examined by PCR analysis [37]. The confirmation of LVFX susceptibility examined by both PCR and culture methods is a future assignment to clarify the reliability of these two methods, which will reveal the clinical impact of LVFX susceptibility on treatment outcome. Second, the antimicrobial resistance patterns differ between Japan and Italy. The rates of LVFX and MNZ resistance are 14.2% in this study and 0–28% as reported in Japan [19,20,22], whilst 9% and more than 40% in Italy, respectively [23,30,33]. LVFX resistance is more common in Japan compared to Italy. In contrast, MNZ resistance is relatively uncommon in Japan, which is characteristic across the world. This might bring about quite a high cure rate of MNZbased therapy in Japan. However, the previous Japanese trial showed a similar high cure rate (88%) by using the same regimen in a patient series harbouring a high MNZ-resistant rate (24.4%) [20]. This may indicate that the high performance of such regimen is related to the Japanese population rather than bacterial strain characteristics. In other words, the possible differences may exist between Western and Japanese patients. Finally, LVFX–AMOX–PPI regimen as second-line therapy may depend on treatment duration rather than LVFX dosage. Indeed, when retreatment with this regimen has been administered for 10 days the cure rates were 70, 83 and 94% [25,30,32], whilst the same therapy administered for 7 days achieved eradication rate ranging from 63 to 70% [24,31]. The possible use of 10-day treatment with LVFX as alternative rescue regimen can be suggested, but needs to be further investigated. In conclusion, MNZ-based triple therapy is a better choice for second-line therapy than LVFX-based triple therapy in the Japanese population. LVFX-based therapy may be an alternative option in the case of MNZ resistance or thirdline therapy. The key to successful rescue therapy with LVFX–AMOX–PPI regimen may be treatment duration, not LVFX dosage. Conflict of interest statement None declared.
List of abbreviations AMOX, amoxicillin; CAM, clarithromycin; CI, confidence intervals; H. pylori, Helicobacter pylori; ITT, intention-to-treat; LPZ, lansoprazole; LVFX, levofloxacin; MNZ, metronidazole; PP, per protocol; PPI, proton pump inhibitor.
Acknowledgments The authors express great thanks to Satomi Shigeta and Ken Yoshimura for excellent assistance. References [1] European Helicobacter pylori Study Group. Current European concepts in the management of Helicobacter pylori infection. The Maastricht Consensus Report. Gut 1997;41:8–13. [2] The Report of the Digestive Health Initiative International Update Conference on Helicobacter pylori. Gastroenterology 1997;113:S4–8. [3] Lam SK, Talley NJ. Report of the 1997 Asia Pacific Consensus Conference on the management of Helicobacter pylori infection. J Gastoroenterol Hepatol 1998;13:1–12. [4] Lind T, Megraud F, Unge P, Bayerdorffer E, O’morain C, Spiller R, et al. The MACH2 study: role of omeprazole in eradication of Helicobacter pylori with 1-week triple therapies. Gastroenterology 1999;116:248–53. [5] Asaka M, Sugiyama T, Kato M, Satoh K, Kuwayama H, Fukuda Y, et al. A multicenter, double-blind study on triple therapy with lansoprazole, amoxicillin and clarithromycin for eradication of Helicobacter pylori in Japanese peptic ulcer patients. Helicobacter 2001;6:254–61. [6] Perri F, Villani MR, Festa V, Quitadamo M, Andriulli A. Predictors of failure of Helicobacter pylori eradication with the standard ’Maastricht triple therapy. Aliment Pharmacol Ther 2001;15:1023–9. [7] Della Monica P, Lavagna A, Masoero G, Lombardo L, Crocella L, Pera A. Effectiveness of Helicobacter pylori eradication treatments in a primary care setting in Italy. Aliment Pharmacol Ther 2002;16:1269–75. [8] Gumurdulu Y, Serin E, Ozer B, Kayaselcuk F, Ozsahin K, Cosar AM, et al. Low eradication rate of Helicobacter pylori with triple 7–14 days and quadriple therapy in Turkey. World J Gastroenterol 2004;10:668–71. [9] Osato MS, Reddy R, Reddy SG, Penland RL, Malaty HM, Graham DY. Pattern of primary resistance of Helicobacter pylori to metronidazole or clarithromycin in the United States. Arch Intern Med 2001;161:1217–20. [10] Poon SK, Chang CS, Su J, Lai CH, Yang CC, Chen GH, et al. Primary resistance to antibiotics and its clinical impact on the efficacy of Helicobacter pylori lansoprazole-based triple therapies. Aliment Pharmacol Ther 2002;16:291–6. [11] Megraud F, Lamouliatte H. Review article: the treatment of refractory Helicobacter pylori infection. Aliment Pharmacol Ther 2003;17:1333–43. [12] Adamek RJ, Suerbaum S, Pfaffenbach B, Opferkuch W. Primary and acquired Helicobacter pylori resistance to clarithromycin, metronidazole, and amoxicillin-influence on treatment outcome. Am J Gastroenterol 1998;93:386–9. [13] Toracchio S, Marzio L. Primary and secondary antibiotic resistance of Helicobacter pylori strains isolated in central Italy during the years 1998–2002. Dig Liver Dis 2003;35:541–5.
Y. Matsumoto et al. / Digestive and Liver Disease 37 (2005) 821–825 [14] Graham DY, Lew GM, Malaty HM, Evans DG, Evans Jr DJ, Klein PD, et al. Factors influencing the eradication of Helicobacter pylori with triple therapy. Gastroenterology 1992;102:493– 6. [15] Asaka M, Satoh K, Sugano K, Sugiyama T, Takahashi S, Fukuda Y, et al. A multicenter, double-blind study on triple therapy with lansoprazole, amoxicillin and clarithromycin for eradication of Helicobacter pylori in Japanese peptic ulcer patients. Helicobacter 2001;6:177–86. [16] Miwa H, Misawa H, Yamada T, Nagahara A, Ohtaka K, Sato N. Clarithromycin resistance, but not CYP2C-19 polymorphism, has a major impact on treatment success in 7-day treatment regimen for cure of H. pylori infection: a multiple logistic regression analysis. Dig Dis Sci 2001;46:2445–50. [17] Miki I, Aoyama N, Sakai T, Shirasaka D, Wambura CM, Maekawa S, et al. Impact of clarithromycin resistance and CYP2C19 genetic polymorphism on treatment efficacy of Helicobacter pylori infection with lansoprazole- or rabeprazole-based triple therapy in Japan. Eur J Gastroenterol Hepatol 2003;15:27–33. [18] Perez Aldana L, Kato M, Nakagawa S, Kawarasaki M, Nagasako T, Mizushima T, et al. The relationship between consumption of antimicrobial agents and the prevalence of primary Helicobacter pylori resistance. Helicobacter 2002;7:306–9. [19] Shimoyama T, Fukuda S, Mikami T, Fukushi M, Munakata A. Efficacy of metronidazole for the treatment of clarithromycin-resistant Helicobacter pylori infection in a Japanese population. J Gastroenterol 2004;39:927–30. [20] Murakami K, Sato R, Okimoto T, Nasu M, Fujioka T, Kodama M, et al. Efficacy of triple therapy comprising rabeprazole, amoxicillin and metronidazole for second-line Helicobacter pylori eradication in Japan, and the influence of metronidazole resistance. Aliment Pharmacol Ther 2003;17:119–23. [21] Hiyama T, Tanaka S, Masuda H, Shima H, Kose K, Tuncel H, et al. Prevalence of Helicobacter pylori resistance to clarithromycin and metronidazole determined by 23S ribosomal RNA and rdxA gene analyses in Hiroshima, Japan. J Gastroenterol Hepatol 2003;18:1202–7. [22] Isomoto H, Inoue K, Furusu H, Enjoji A, Fujimoto C, Yamakawa M, et al. High-dose rabeprazole–amoxicillin versus rabeprazole–amoxicillin–metronidazole as second-line treatment after failure of the Japanese standard regimen for Helicobacter pylori infection. Aliment Pharmacol Ther 2003;18:101–7. [23] Branca G, Spanu T, Cammarota G, Schito AM, Gasbarrini A, Gasbarrini GB, et al. High levels of dual resistance to clarithromycin and metronidazole and in vitro activity of levofloxacin against Helicobacter pylori isolates from patients after failure of therapy. Int J Antimicrob Agents 2004;24:433–8. [24] Watanabe Y, Aoyama N, Shirasaka D, Maekawa S, Kuroda K, Miki I, et al. Levofloxacin based triple therapy as a second-line treatment after failure of Helicobacter pylori eradication with standard triple therapy. Dig Liver Dis 2003;35:711–5.
825
[25] Nista EC, Candelli M, Cremonini F, Cazzato IA, Di Caro S, Gabrielli M, et al. Levofloxacin-based triple therapy vs. quadruple therapy in second-line Helicobacter pylori treatment: a randomized trial. Aliment Pharmacol Ther 2003;18:627–33. [26] Ohara S, Kato M, Asaka M, Toyota T. Studies of 13 C-urea breath test for diagnosis of Helicobacter pylori infection in Japan. J Gastroenterol 1998;33:6–13. [27] Perri F, Festa V, Clemente R, Quitadamo M, Andriulli A. Rifabutin-based ‘rescue therapy’ for Helicobacter pylori infected patients after failure of standard regimens. Aliment Pharmacol Ther 2000;14:311–6. [28] Perri F, Festa V, Clemente R, Villani MR, Quitadamo M, Caruso N, et al. Randomized study of two “rescue” therapies for Helicobacter pylori-infected patients after failure of standard triple therapies. Am J Gastroenterol 2001;96:58–62. [29] Wong WM, Gu Q, Lam SK, Fung FM, Lai KC, Hu WH, et al. Randomized controlled study of rabeprazole, levofloxacin and rifabutin triple therapy vs. quadruple therapy as second-line treatment for Helicobacter pylori infection. Aliment Pharmacol Ther 2003;17:553–60. [30] Bilardi C, Dulbecco P, Zentilin P, Reglioni S, Iiritano E, Parodi A, et al. A 10-day levofloxacin-based therapy in patients with resistant Helicobacter pylori infection: a controlled trial. Clin Gastroenterol Hepatol 2004;2:997–1002. [31] Perri F, Festa V, Merla A, Barberani F, Pilotto A, Andriulli A. Randomized study of different ‘second-line’ therapies for Helicobacter pylori infection after failure of the standard ‘Maastricht triple therapy’. Aliment Pharmacol Ther 2003;18:815–20. [32] Zullo A, Hassan C, De Francesco V, Lorenzetti R, Marignani M, Angeletti S, et al. A third-line levofloxacin-based rescue therapy for Helicobacter pylori eradication. Dig Liver Dis 2003;35:232–6. [33] Cammarota G, Martino A, Pirozzi G, Cianci R, Branca G, Nista EC, et al. High efficacy of 1-week doxycycline- and amoxicillinbased quadruple regimen in a culture-guided, third-line treatment approach for Helicobacter pylori infection. Aliment Pharmacol Ther 2004;19:789–95. [34] Hachem CY, Clarridge JE, Reddy R, Flamm R, Evans DG, Tanaka SK, et al. Antimicrobial susceptibility testing of Helicobacter pylori. Comparison of E-test, broth microdilution, and disk diffusion for ampicillin, clarithromycin, and metronidazole. Diagn Microbiol Infect Dis 1996;24:37–41. [35] Osato MS, Reddy R, Reddy SG, Penland RL, Graham DY. Comparison of the E test and the NCCLS-approved agar dilution method to detect metronidazole and clarithromycin resistant Helicobacter pylori. Int J Antimicrob Agents 2001;17:39–44. [36] Masaoka T, Suzuki H, Kurabayashi K, Kamiya AG, Ishii H. Second-line treatment of Helicobacter pylori infection after dilution agar methods and PCR-RFLP analysis. Aliment Pharmacol Ther 2004;20:68–73. [37] Fujimura S, Kato S, Iinuma K, Watanabe A. In vitro activity of fluoroquinolone and the gyrA gene mutation in Helicobacter pylori strains isolated from children. J Med Microbiol 2004;53:1019–22.
Alimentary Tract
Levofloxacin- versus metronidazole-based rescue therapy for H. pylori infection in Japan Y. Matsumoto a , I. Miki a , N. Aoyama b,∗ , D. Shirasaka a , Y. Watanabe a , Y. Morita a , M. Toyoda a , T. Mitani a , H. Miyachi a , T. Tamura a , M. Kasuga a a
b
Division of Diabetes, Digestive and Kidney Diseases, Department of Clinical Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan Department of Endoscopy, Kobe University Graduate School of Medicine, 7-5-1, Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan Received 26 April 2005; accepted 14 June 2005 Available online 22 July 2005
Abstract Background. The ideal second-line treatment regimens for Helicobacter pylori infection may differ between the areas, countries and races. Aim. The aim was to confirm which was the better regimen for second-line therapy after treatment failure with a standard triple therapy in Japan, a high dosage of levofloxacin- or metronidazole-based therapy. Patients. Sixty outpatients with persistent H. pylori infection after a standard triple therapy were enrolled in this prospective, open-label and randomised trial. Methods. The subjects were randomly administered levofloxacin (300 mg b.d.)- or metronidazole (500 mg b.d.)-based therapy with lansoprazole (30 mg b.d.) and amoxicillin (1000 mg b.d.) for 7 days, and the cure rates and side effects were analysed. Antimicrobial susceptibility was also examined before second-line therapy using the E-test. Results. Good compliance was obtained without severe side effects in both the groups except for two patients. The cure rates, expressed as intention-to-treat and per-protocol analyses, respectively, were 70.0 and 72.4% in the levofloxacin group, and 96.7 and 100% in the metronidazole group. Each regimen often overcame even clarithromycin-resistant strains. Conclusion. Metronidazole-based triple therapy is recommended as second-line therapy in Japan, and levofloxacin-based therapy can be an alternative treatment option. © 2005 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. Keywords: H. pylori; Levofloxacin; Metronidazole; Second-line
1. Introduction Proton pump inhibitor (PPI)-based triple therapy, containing two antibiotics among amoxicillin (AMOX), clarithromycin (CAM) and metronidazole (MNZ), is widely considered the golden standard for first-line therapy of H. pylori infection [1–3]. However, the therapeutic efficacy varies widely, ranging 25–95% [4–8]. Recent years have witnessed a decrease in its cure rate due to the progres∗ Corresponding author. Tel.: +81 78 382 5863/6570; fax: +81 78 382 6579. E-mail address: [email protected] (N. Aoyama).
sion of antimicrobial resistance and currently, individuals with refractory H. pylori infection cannot be ignored [9–13]. Reportedly, poor compliance and antimicrobial resistance, in particular CAM or MNZ resistance, are considered some of the most important factors of treatment failure [9,11,14]. It is, therefore, an urgent issue to develop an appropriate eradication strategy. The best rescue treatment regimens should differ among countries and races because of differences in antimicrobial resistance patterns. In Japan, PPI with AMOX and CAM is recommended by the Japanese Helicobacter Society as first-line therapy against H. pylori infection [15], and the treatment efficacy of this regimen mainly depends on CAM
1590-8658/$30 © 2005 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dld.2005.06.002
822
Y. Matsumoto et al. / Digestive and Liver Disease 37 (2005) 821–825
susceptibility [16,17]. In case of treatment failure, no fixed regimen has been established for second-line therapy in Japan. The ideal rescue regimen should consist of short-term and simple drug combinations using susceptible agents with good compliance and little toxicity. Of several rescue regimens proposed in various countries, the combination of PPI, AMOX and MNZ is often administered as second-line treatment in Japan, where MNZ resistance is relatively uncommon compared to worldwide tendency [18–22]. The cure rate of the 7-day PPI–AMOX–MNZ regimen as second-line therapy is around 90% in Japan [19,20,22]. This regimen has the advantage of compliance because of its simple and short-term treatment. However, MNZ resistance may also be widespread in Japan due to its increased consumption, as with CAM resistance. Furthermore, dual resistance to CAM and MNZ is often detected [21,23]. Therefore, it is important to find a new effective antibiotic with low toxicity instead of CAM or MNZ. Of several candidates a new fluoroquinolone, levofloxacin (LVFX), is considered an attractive agent for second-line treatment due to its excellent activity against H. pylori [23,24]. Our previous study revealed that LVFX has strong activity even against CAM-resistant H. pylori strains in vitro, and the efficacy and safety of LVFX 400 mg-based therapy were confirmed as second-line therapy, showing a successful eradication rate of 70% [24]. According to an Italian trial, LVFX 500 mg-based therapy showed a high re-eradication rate of 94% [25]. Focusing on the difference in LVFX dosage, we hypothesised that a high dosage of LVFX-based therapy could induce increased efficacy, and that this regimen might compare with MNZ-based regimen in efficacy and toxicity in the Japanese population. This study was carried out to confirm which was the ideal second-line regimen for refractory H. pylori infection to a standard triple therapy in Japan, a high dosage of LVFX-based therapy or MNZ-based therapy.
2. Patients and methods This study was carried out in a prospective, open-label, randomised, parallel group fashion. The study subjects were 60 consecutive outpatients with persistent H. pylori infection after a standard triple therapy (gastric ulcer, n = 26; duodenal ulcer, n = 14; gastroduodenal ulcer, n = 4; gastritis, n = 16; 36 males and 24 females; mean age, 53.1 ± 13.7). These patients were enrolled according to the eligibility criteria (age greater than 20 years and younger than 70 years) and the exclusion criteria (patients who had been taking aspirin, other non-steroidal anti-inflammatory drugs, known drug allergy to the study drugs, gastric cancer, severe concomitant disease and previous gastric surgery). All the patients underwent endoscopy before first-line therapy to confirm gastroduodenal disease and received 7-day PPI–AMOX–CAM therapy as first-line therapy. H. pylori status was confirmed by a 13 C-urea breath test before the trial. The cut-off value was
Table 1 Clinical characteristics of LVFX and MNZ groups LVFX group (n = 30)
MNZ group (n = 30)
Age (years) Male/female
50.8 ± 13.5 17/13
52.0 ± 13.0 19/11
Disease GU DU GDU Gastritis
15 6 2 7
11 8 2 9
First-line regimen LPZ–AMOX–CAM OPZ–AMOX–CAM RPZ–AMOX–CAM
27 1 2
28 1 1
Smoking/non-smoking Drinking/non-drinking AMOX-S/-R/unknown CAM-S/-R/unknown LVFX-S/-R/unknown MNZ-S/-R/unknown
8/22 13/17 17/0/13 5/12/13 15/2/13 15/2/13
14/16 16/14 18/0/12 9/9/12 15/3/12 17/1/12
GU, gastric ulcer; DU, duodenal ulcer; GDU, gastroduodenal ulcer; LPZ, lansoprazole 30 mg b.d.; AMOX, amoxicillin 750 mg b.d.; CAM, clarithromycin 200 mg or 400 mg b.d.; RPZ, rabeprazole 20 mg b.d.; OPZ, omeprazole 20 mg b.d.; -S, suspeptible; -R, resistant.
set at 2.5‰ 20 min after taking 100 mg 13 C-urea [26]. Follow-up endoscopy was performed for a biopsy to examine the antimicrobial susceptibility of H. pylori before secondline treatment when agreement was obtained. The antimicrobial susceptibility testing was assessed using the E-test (AB Biodisk, Solna, Sweden) as previously described [17]. AMOX, CAM, LVFX and MNZ resistance were defined as the minimum inhibitory concentration of >8, >1, >1 and >8 g/ml, respectively. The result of the susceptibility testing was not available before the evaluation of the secondline treatment outcome. After assessment for refractory H. pylori infection to standard PPI-based triple therapy, these patients were randomly assigned by computer-generated random numbers to receive one of the following treatments: (a) 300 mg LVFX b.d., 20 mg lansoprazole (LPZ) b.d. and 1000 mg AMOX b.d. for 7 days (LVFX group; n = 30), or (b) 500 mg MNZ b.d., 20 mg lansoprazole b.d. and 1000 mg AMOX b.d. for 7 days (MNZ group; n = 30). The clinical characteristics of each group are shown in Table 1. Compliance and side effects were documented by a detailed questionnaire sheet during treatment. The patients were advised to fill in the sheet every day, which consisted of several columns expressing the drug intake time and the grade (mild–moderate–severe) of possible side effects (nausea, vomiting, abdominal discomfort, soft stool, diarrhoea, constipation, loss of appetite, taste disturbance, headache, vertigo, general fatigue, skin rash and others). A 13 C-urea breath test was performed 8 weeks after the completion of second-line therapy. Treatment efficacy was expressed using both intentionto-treat (ITT) and per-protocol (PP) analyses. The 95%
Y. Matsumoto et al. / Digestive and Liver Disease 37 (2005) 821–825 Table 2 Side effects of LVFX and MNZ groups
823
Table 3 PP cure rate according to antimicrobial resistance pattern
LVFX group
MNZ group
Diarrhoea/soft stool Nausea General fatigue Abdominal discomfort Skin rash
3 1 1 1 1
6 3 1 1 0
Total episodes Total patients (%)
7 6 (20.0)
11 8 (26.7)
LVFX, levofloxacin; MNZ, metronidazole.
confidence intervals (CI) were also calculated. The Fisher’s exact test or χ2 -test were used to compare the differences in clinical characteristics, side effects and eradication rates of each group. A p-value less than 0.05 was considered significant. This study was approved by the Interstitial Review Board of Kobe University Hospital. Written informed consent was obtained from all subjects before their enrollment in this study.
LVFX group CAM-S, LVFX-S CAM-S, LVFX-R CAM-R, LVFX-S CAM-R, LVFX-R Unknown Total
3/4 (75.0%) – 6/10 (60.0%) 1/2 (50.0%) 11/13 (84.6%) 21/29 (72.4%)
MNZ group CAM-S, MNZ-S CAM-S, MNZ-R CAM-R, MNZ-S CAM-R, MNZ-R Unknown Total
8/8 (100%) 1/1 (100%) 8/8 (100%) – 12/12 (100%) 29/29 (100%)
LVFX, levofloxacin; MNZ, metronidazole; CAM, clarithromycin; -S, susceptible; -R, resistant.
ture in the LVFX group, and all the patients without culture in the MNZ group showed successful eradication.
4. Discussion 3. Results Sixty patients were enrolled in this trial and were divided into two groups. The two groups were well matched with regard to demographic and clinical characteristics (Table 1). All patients completed the therapeutic regimens except for two: one patient in the LVFX group dropped out because of mild skin rash, and another in the MNZ group because of severe diarrhoea. Except for these two, the side effects were very mild in both the groups, and the patients maintained good compliance. The overall prevalence of side effects was similar and no specific adverse effects were found in either regimen (Table 2). The cure rate, expressed as ITT and PP analyses, respectively, was 70.0% (95% CI = 45–95) and 72.4% (95% CI = 56–89) in the LVFX group and 96.7% (95% CI = 90–100) and 100% in the MNZ group. The treatment outcome showed a significantly high cure rate in the MNZ group compared to that in the LVFX group (p = 0.006 in ITT analysis and p = 0.002 in PP analysis). H. pylori susceptibility testing was performed in 38 patients before second-line treatment and the results of the E-test were successfully obtained in 35 patients, including the two patients who dropped out. The antimicrobial resistance pattern of each group was shown in Table 1. The overall CAM, LVFX and MNZ resistance were found in 21 (60.0%), five (14.2%) and three (8.6%) out of 35 examined, respectively. AMOX resistance was not found. Table 3 shows the PP cure rate according to each antimicrobial resistance pattern. MNZ-based therapy could eradicate all the strains, even MNZ- and CAM-resistant strains. On the other hand, LVFX-based therapy could eradicate seven out of 12 (58.3%) CAM-resistant strains. Eleven out of 13 patients without cul-
This trial was conducted in order to establish a simple and short-term regimen with effective and non-toxic agents for an ideal second-line treatment for H. pylori infection in Japan. Many investigators have identified attractive agents, such as rifabutin and furazolidone; however, these have limited use in consideration of toxicity, specific utility and cost benefit [27–29]. Of several other candidates, we focused on LVFX. PPI with LVFX and AMOX is a promising regimen for the salvage therapy [24,25,30–32]. These studies support that LVFX-based therapy is an attractive treatment option; however, only few data are available concerning LVFX susceptibility in the clinical trials [24,33]. Our previous trial using LVFX 400 mg-based regimen did not show entirely satisfactory re-eradication rate [24]. Therefore, this trial was conducted using a high dosage of LVFX (600 mg) in order to approach the high cure rate of LVFX 500 mg-based regimens reported by Italian groups [25]. In contrast to our expectation, however, the cure rate was not higher than that of our previous trial with LVFX 400 mg (72.4% versus 69.7% according to PP analysis) [24], or the Italian groups’ report (72.4% versus 94% according to PP analysis) [25] and this regimen was not superior to MNZ-based triple therapy (72.4% versus 100% according to PP analysis), although it may not be appropriate to compare the results of different trials. In addition, MNZ has the advantage of low cost compared to LVFX. One tablet of MNZ and LVFX costs 42 and 212 Japanese yen, respectively. LVFX resistance was not rare in this trial(14.2%). LVFXbased triple therapy could often eradicate even CAMresistant strains. However, this regimen could not sometimes eradicate even LVFX-susceptible strains (cure rate of only 60.0% according to PP analysis), although LVFX showed excellent activity in vitro [23,24]. It cannot be denied that the cure rate of LVFX-based therapy in this trial may be estimated
824
Y. Matsumoto et al. / Digestive and Liver Disease 37 (2005) 821–825
lower than the actual efficacy because of the limited number of patients. The results obtained can be discussed according to the following factors: first, LVFX susceptibility judged by the E-test may not always predict the treatment outcome. As for MNZ, there may be a discrepancy between the E-test and the agar dilution method [34,35]. As for CAM, the results of susceptibility obtained by the culture method may be different from those obtained by the PCR method [36]. In this study, the PP cure rate of LVFX-susceptible strains was only 64.3%, which was lower than that of all the patients treated with LVFX (72.4%). The same content as found in MNZ and CAM may also apply to LVFX. The LVFX-resistance mechanism is screened by DNA gyrase mutations, which can be examined by PCR analysis [37]. The confirmation of LVFX susceptibility examined by both PCR and culture methods is a future assignment to clarify the reliability of these two methods, which will reveal the clinical impact of LVFX susceptibility on treatment outcome. Second, the antimicrobial resistance patterns differ between Japan and Italy. The rates of LVFX and MNZ resistance are 14.2% in this study and 0–28% as reported in Japan [19,20,22], whilst 9% and more than 40% in Italy, respectively [23,30,33]. LVFX resistance is more common in Japan compared to Italy. In contrast, MNZ resistance is relatively uncommon in Japan, which is characteristic across the world. This might bring about quite a high cure rate of MNZbased therapy in Japan. However, the previous Japanese trial showed a similar high cure rate (88%) by using the same regimen in a patient series harbouring a high MNZ-resistant rate (24.4%) [20]. This may indicate that the high performance of such regimen is related to the Japanese population rather than bacterial strain characteristics. In other words, the possible differences may exist between Western and Japanese patients. Finally, LVFX–AMOX–PPI regimen as second-line therapy may depend on treatment duration rather than LVFX dosage. Indeed, when retreatment with this regimen has been administered for 10 days the cure rates were 70, 83 and 94% [25,30,32], whilst the same therapy administered for 7 days achieved eradication rate ranging from 63 to 70% [24,31]. The possible use of 10-day treatment with LVFX as alternative rescue regimen can be suggested, but needs to be further investigated. In conclusion, MNZ-based triple therapy is a better choice for second-line therapy than LVFX-based triple therapy in the Japanese population. LVFX-based therapy may be an alternative option in the case of MNZ resistance or thirdline therapy. The key to successful rescue therapy with LVFX–AMOX–PPI regimen may be treatment duration, not LVFX dosage. Conflict of interest statement None declared.
List of abbreviations AMOX, amoxicillin; CAM, clarithromycin; CI, confidence intervals; H. pylori, Helicobacter pylori; ITT, intention-to-treat; LPZ, lansoprazole; LVFX, levofloxacin; MNZ, metronidazole; PP, per protocol; PPI, proton pump inhibitor.
Acknowledgments The authors express great thanks to Satomi Shigeta and Ken Yoshimura for excellent assistance. References [1] European Helicobacter pylori Study Group. Current European concepts in the management of Helicobacter pylori infection. The Maastricht Consensus Report. Gut 1997;41:8–13. [2] The Report of the Digestive Health Initiative International Update Conference on Helicobacter pylori. Gastroenterology 1997;113:S4–8. [3] Lam SK, Talley NJ. Report of the 1997 Asia Pacific Consensus Conference on the management of Helicobacter pylori infection. J Gastoroenterol Hepatol 1998;13:1–12. [4] Lind T, Megraud F, Unge P, Bayerdorffer E, O’morain C, Spiller R, et al. The MACH2 study: role of omeprazole in eradication of Helicobacter pylori with 1-week triple therapies. Gastroenterology 1999;116:248–53. [5] Asaka M, Sugiyama T, Kato M, Satoh K, Kuwayama H, Fukuda Y, et al. A multicenter, double-blind study on triple therapy with lansoprazole, amoxicillin and clarithromycin for eradication of Helicobacter pylori in Japanese peptic ulcer patients. Helicobacter 2001;6:254–61. [6] Perri F, Villani MR, Festa V, Quitadamo M, Andriulli A. Predictors of failure of Helicobacter pylori eradication with the standard ’Maastricht triple therapy. Aliment Pharmacol Ther 2001;15:1023–9. [7] Della Monica P, Lavagna A, Masoero G, Lombardo L, Crocella L, Pera A. Effectiveness of Helicobacter pylori eradication treatments in a primary care setting in Italy. Aliment Pharmacol Ther 2002;16:1269–75. [8] Gumurdulu Y, Serin E, Ozer B, Kayaselcuk F, Ozsahin K, Cosar AM, et al. Low eradication rate of Helicobacter pylori with triple 7–14 days and quadriple therapy in Turkey. World J Gastroenterol 2004;10:668–71. [9] Osato MS, Reddy R, Reddy SG, Penland RL, Malaty HM, Graham DY. Pattern of primary resistance of Helicobacter pylori to metronidazole or clarithromycin in the United States. Arch Intern Med 2001;161:1217–20. [10] Poon SK, Chang CS, Su J, Lai CH, Yang CC, Chen GH, et al. Primary resistance to antibiotics and its clinical impact on the efficacy of Helicobacter pylori lansoprazole-based triple therapies. Aliment Pharmacol Ther 2002;16:291–6. [11] Megraud F, Lamouliatte H. Review article: the treatment of refractory Helicobacter pylori infection. Aliment Pharmacol Ther 2003;17:1333–43. [12] Adamek RJ, Suerbaum S, Pfaffenbach B, Opferkuch W. Primary and acquired Helicobacter pylori resistance to clarithromycin, metronidazole, and amoxicillin-influence on treatment outcome. Am J Gastroenterol 1998;93:386–9. [13] Toracchio S, Marzio L. Primary and secondary antibiotic resistance of Helicobacter pylori strains isolated in central Italy during the years 1998–2002. Dig Liver Dis 2003;35:541–5.
Y. Matsumoto et al. / Digestive and Liver Disease 37 (2005) 821–825 [14] Graham DY, Lew GM, Malaty HM, Evans DG, Evans Jr DJ, Klein PD, et al. Factors influencing the eradication of Helicobacter pylori with triple therapy. Gastroenterology 1992;102:493– 6. [15] Asaka M, Satoh K, Sugano K, Sugiyama T, Takahashi S, Fukuda Y, et al. A multicenter, double-blind study on triple therapy with lansoprazole, amoxicillin and clarithromycin for eradication of Helicobacter pylori in Japanese peptic ulcer patients. Helicobacter 2001;6:177–86. [16] Miwa H, Misawa H, Yamada T, Nagahara A, Ohtaka K, Sato N. Clarithromycin resistance, but not CYP2C-19 polymorphism, has a major impact on treatment success in 7-day treatment regimen for cure of H. pylori infection: a multiple logistic regression analysis. Dig Dis Sci 2001;46:2445–50. [17] Miki I, Aoyama N, Sakai T, Shirasaka D, Wambura CM, Maekawa S, et al. Impact of clarithromycin resistance and CYP2C19 genetic polymorphism on treatment efficacy of Helicobacter pylori infection with lansoprazole- or rabeprazole-based triple therapy in Japan. Eur J Gastroenterol Hepatol 2003;15:27–33. [18] Perez Aldana L, Kato M, Nakagawa S, Kawarasaki M, Nagasako T, Mizushima T, et al. The relationship between consumption of antimicrobial agents and the prevalence of primary Helicobacter pylori resistance. Helicobacter 2002;7:306–9. [19] Shimoyama T, Fukuda S, Mikami T, Fukushi M, Munakata A. Efficacy of metronidazole for the treatment of clarithromycin-resistant Helicobacter pylori infection in a Japanese population. J Gastroenterol 2004;39:927–30. [20] Murakami K, Sato R, Okimoto T, Nasu M, Fujioka T, Kodama M, et al. Efficacy of triple therapy comprising rabeprazole, amoxicillin and metronidazole for second-line Helicobacter pylori eradication in Japan, and the influence of metronidazole resistance. Aliment Pharmacol Ther 2003;17:119–23. [21] Hiyama T, Tanaka S, Masuda H, Shima H, Kose K, Tuncel H, et al. Prevalence of Helicobacter pylori resistance to clarithromycin and metronidazole determined by 23S ribosomal RNA and rdxA gene analyses in Hiroshima, Japan. J Gastroenterol Hepatol 2003;18:1202–7. [22] Isomoto H, Inoue K, Furusu H, Enjoji A, Fujimoto C, Yamakawa M, et al. High-dose rabeprazole–amoxicillin versus rabeprazole–amoxicillin–metronidazole as second-line treatment after failure of the Japanese standard regimen for Helicobacter pylori infection. Aliment Pharmacol Ther 2003;18:101–7. [23] Branca G, Spanu T, Cammarota G, Schito AM, Gasbarrini A, Gasbarrini GB, et al. High levels of dual resistance to clarithromycin and metronidazole and in vitro activity of levofloxacin against Helicobacter pylori isolates from patients after failure of therapy. Int J Antimicrob Agents 2004;24:433–8. [24] Watanabe Y, Aoyama N, Shirasaka D, Maekawa S, Kuroda K, Miki I, et al. Levofloxacin based triple therapy as a second-line treatment after failure of Helicobacter pylori eradication with standard triple therapy. Dig Liver Dis 2003;35:711–5.
825
[25] Nista EC, Candelli M, Cremonini F, Cazzato IA, Di Caro S, Gabrielli M, et al. Levofloxacin-based triple therapy vs. quadruple therapy in second-line Helicobacter pylori treatment: a randomized trial. Aliment Pharmacol Ther 2003;18:627–33. [26] Ohara S, Kato M, Asaka M, Toyota T. Studies of 13 C-urea breath test for diagnosis of Helicobacter pylori infection in Japan. J Gastroenterol 1998;33:6–13. [27] Perri F, Festa V, Clemente R, Quitadamo M, Andriulli A. Rifabutin-based ‘rescue therapy’ for Helicobacter pylori infected patients after failure of standard regimens. Aliment Pharmacol Ther 2000;14:311–6. [28] Perri F, Festa V, Clemente R, Villani MR, Quitadamo M, Caruso N, et al. Randomized study of two “rescue” therapies for Helicobacter pylori-infected patients after failure of standard triple therapies. Am J Gastroenterol 2001;96:58–62. [29] Wong WM, Gu Q, Lam SK, Fung FM, Lai KC, Hu WH, et al. Randomized controlled study of rabeprazole, levofloxacin and rifabutin triple therapy vs. quadruple therapy as second-line treatment for Helicobacter pylori infection. Aliment Pharmacol Ther 2003;17:553–60. [30] Bilardi C, Dulbecco P, Zentilin P, Reglioni S, Iiritano E, Parodi A, et al. A 10-day levofloxacin-based therapy in patients with resistant Helicobacter pylori infection: a controlled trial. Clin Gastroenterol Hepatol 2004;2:997–1002. [31] Perri F, Festa V, Merla A, Barberani F, Pilotto A, Andriulli A. Randomized study of different ‘second-line’ therapies for Helicobacter pylori infection after failure of the standard ‘Maastricht triple therapy’. Aliment Pharmacol Ther 2003;18:815–20. [32] Zullo A, Hassan C, De Francesco V, Lorenzetti R, Marignani M, Angeletti S, et al. A third-line levofloxacin-based rescue therapy for Helicobacter pylori eradication. Dig Liver Dis 2003;35:232–6. [33] Cammarota G, Martino A, Pirozzi G, Cianci R, Branca G, Nista EC, et al. High efficacy of 1-week doxycycline- and amoxicillinbased quadruple regimen in a culture-guided, third-line treatment approach for Helicobacter pylori infection. Aliment Pharmacol Ther 2004;19:789–95. [34] Hachem CY, Clarridge JE, Reddy R, Flamm R, Evans DG, Tanaka SK, et al. Antimicrobial susceptibility testing of Helicobacter pylori. Comparison of E-test, broth microdilution, and disk diffusion for ampicillin, clarithromycin, and metronidazole. Diagn Microbiol Infect Dis 1996;24:37–41. [35] Osato MS, Reddy R, Reddy SG, Penland RL, Graham DY. Comparison of the E test and the NCCLS-approved agar dilution method to detect metronidazole and clarithromycin resistant Helicobacter pylori. Int J Antimicrob Agents 2001;17:39–44. [36] Masaoka T, Suzuki H, Kurabayashi K, Kamiya AG, Ishii H. Second-line treatment of Helicobacter pylori infection after dilution agar methods and PCR-RFLP analysis. Aliment Pharmacol Ther 2004;20:68–73. [37] Fujimura S, Kato S, Iinuma K, Watanabe A. In vitro activity of fluoroquinolone and the gyrA gene mutation in Helicobacter pylori strains isolated from children. J Med Microbiol 2004;53:1019–22.