Metronidazole resistance and virulence factors in Helicobacter pylori as markers for treatment failure in a paediatric population

Transcription

1 FEMS Immunology and Medical Microbiology 24 (1999) 183^188 Metronidazole resistance and virulence factors in Helicobacter pylori as markers for treatment failure in a paediatric population Abstract Manuel Löpez-Brea a; *, Maria Josefa Mart nez b, Diego Domingo a, Isabel Sänchez a, Teresa Alarcön a a Department of Clinical Microbiology, Hospital Universitario de la Princesa, Madrid, Spain b Gastroenterology Unit, Hospital del Nin o Jesüs, Madrid, Spain Received 25 November 1998; accepted 18 February 1999 The eradication rate obtained using the classical triple therapy containing metronidazole, amoxicillin and bismuth citrate, was determined in 57 paediatric patients with digestive disorders, according to the susceptibility to metronidazole of the Helicobacter pylori strains (determined by agar dilution) and the caga and vaca status (determined by PCR). Eradication was obtained in 38 out of 43 patients (88.3%) infected by H. pylori with metronidazole MIC 92 mgl 31, in 3 out of 6 patients (50%) when MIC was 4^8 mg l 31 and in 4 out of 8 patients (50%) when MIC was s 8mgl 31. Among patients infected with caga+ and caga3 strains an eradication rate of 75% (6/8) and 75% (18/24) was found, and 50% (3/6) and 80% (21/26) among vaca s1- and vaca s2-infected subjects (P s 0.05). H. pylori eradication depends on the susceptibility of the strain to metronidazole, being higher in patients infected with susceptible H. pylori. However, according to our data the caga or vaca status was not an important factor in treatment failure in the eradication of H. pylori. ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords: Helicobacter pylori; Metronidazole; Resistance; Virulence factor; Eradication; Pediatric 1. Introduction Helicobacter pylori infection is now recognised as a causal factor in the pathogenesis of gastritis and peptic ulcer and appropriate treatment and successful eradication of the bacteria cure both diseases [1]. Although peptic ulcer is not frequent in children, they could present active or non-active gastritis and other symptoms, such as recurrent abdominal pain, vomiting, weight loss, etc. associated with H. pylori * Corresponding author. infection [2]. After H. pylori eradication most of the symptomatology disappears. H. pylori is susceptible to most antimicrobial agents in vitro, however, in vivo eradication could be di cult. Based on in vitro and clinical studies the agents potentially useful in H. pylori eradication include amoxicillin, tetracycline, metronidazole, clarithromycin, bismuth salts and proton pump inhibitors. Metronidazole is one of the most used antibiotics, frequently as a triple therapy with bismuth salts or proton pump inhibitors and amoxicillin, tetracycline or clarithromycin. Failure to eradicate relates to a number of potential problems, / 99 / $20.00 ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S (99)

2 184 M. Löpez-Brea et al. / FEMS Immunology and Medical Microbiology 24 (1999) 183^188 including patient compliance [3], side e ects [4], natural and acquired bacterial resistance to antibiotics [5^7] or poor delivery of drugs to the site of action [8]. Infection with metronidazole-resistant H. pylori is considered a risk factor for treatment failure in adults, although this fact is less known in paediatric population. Several virulence factors have been described in H. pylori, the presence of the caga gene or the vacuolating toxin being the most studied. Many studies exist where a correlation between the presence of these factors and the virulence of strains is found [9], although in others a very high prevalence of these markers is found in H. pylori from the general adult population and no correlation exists [10]. Some reports suggest that the eradication of the bacteria from the gastric mucosa could be related to caga status, it being more di cult to eradicate the caganegative H. pylori strains [11]. The aim of this study was to determine the eradication rate obtained using the classical triple therapy containing metronidazole, amoxicillin and bismuth citrate, according to the susceptibility to metronidazole and the caga and vaca status of the strains. 2. Materials and methods 2.1. Patients Fifty-seven paediatric patients referred to the Gastroenterology Unit of the Hospital del Nin o Jesüs for digestive disorders were included: 48 patients su ered from gastritis, six from duodenitis and three from duodenal ulcer (medium age was 10 years and 3 months, ranging from 3 to 18 years). An endoscopy was performed to diagnose the disease and four samples for rapid urease test (one sample from antrum), histology (two samples from antrum and corpus) and culture (one sample from antrum) were taken. Serum for serology and a sample for 13 C-UBT were also collected. A patient was considered H. pylori-positive when culture or histology was positive and H. pylorinegative when all tests were negative. Only the culture-positive patients were included in this study. Anti-H. pylori treatment was applied using amoxicillin, metronidazole and bismuth citrate and the follow-up was performed 1, 6 and 12 months after treatment. The follow-up was carried out by serology and endoscopic methods in the rst part of the study; however, when 13 C-UBT was available, the post-treatment endoscopy was avoided. Eradication was considered to have occurred when all methods were negative 1 or 6 months after treatment Culture Biopsies were cultured on selective (Pylori agar, BioMerieux, Lyon, France) and non-selective (Columbia agar supplemented with 7% sheep blood) media and incubated under a microaerophilic atmosphere for at least 8 days. Colonies similar to H. pylori were identi ed by Gram stain morphology (curved Gram-negative rods) and by producing a positive result with urease, catalase and oxidase tests [12]. Susceptibility testing was performed and strains were stored at 380³C in TSB+20% glycerol until used for determination of caga and vaca status Histology The presence of curved Gram-negative rods was determined in the gastric mucosa by Giemsa stain Serology Detection of IgG and IgA was performed by quantitative EIAs (EIA-G and EIA-A Pyloriset; Orion Diagnostica) following the manufacturer's recommendations C-UBT The presence of H. pylori infection was determined with the urea breath test following the European method and positivity established as 13 C detection of v5 per Susceptibility tests Susceptibility to metronidazole was tested with the agar dilution method, using Mueller-Hinton agar plus 7% horse blood containing the antibiotic at different concentrations (two-fold dilutions from 128 to mg l 31 ). Strains were grown in brain heart

3 M. Löpez-Brea et al. / FEMS Immunology and Medical Microbiology 24 (1999) 183^ infusion plus 10% foetal calf serum for 48 h and inoculated on plates using a Steer replicator. The plates were incubated for 3^5 days in a CO 2 incubator. The minimum inhibitory concentration (MIC) was de ned as the lowest concentration of the antimicrobial agent which completely inhibited any visible growth. In all susceptibility tests H. pylori NCTC was used as quality control Determination of virulence markers The presence of the caga gene and the vaca s1 or s2 alleles could be studied in 32 strains. DNA was obtained from a 48-h H. pylori culture by the CTAB reagent method. A 297-bp caga gene fragment was ampli ed by PCR [13]. The vaca s1 and vaca s2 alleles were identi ed by a PCR method [14]. The ampli ed fragment was detected after agarose gel electrophoresis in a UV light transilluminator. 3. Results Eradication was found in 38 out of 43 patients (88.3%) infected by strains with MICs 92 mgl 31, in three out of six patients (50%) when the MIC was 4^8 mg l 31 and in four out of eight patients (50%) when the MIC was s 8mgl 31. The number and percentage of patients with eradication or not according to the metronidazole MIC value are included in Table 1. A 75% (6/8) and 75% (18/24) eradication rate was found among patients infected with caga+ and caga3 clinical isolates, respectively. A 50% (3/6) and 80% (21/26) eradication rate was found among vaca s1- and vaca s2-infected patients, with no statistically signi cant di erences. Eradication was obtained in all patients su ering from duodenal ulcer (3/3), in 77% of patients with gastritis (37/48) and in 50% of patients with duodenitis (3/6). 4. Discussion This study compares the e cacy of the classical anti-h. pylori triple therapy in paediatric patients infected with strains with metronidazole MICs of 92 mgl 31,438 mgl 31 or s 8mgl 31. Eradication was obtained in 88.3% of patients infected with strains with a metronidazole MIC 92 mgl 31, but in only 50% of patients infected with strains with a metronidazole MIC of v4 mg l 31. This means a signi cantly impaired eradication e cacy of triple therapy when the metronidazole MIC is equal to or higher than 4 mg l 31. The prevalence rate of metronidazole resistance among H. pylori strains has been studied worldwide and is variable: higher in developing than in developed countries [15]. Resistance rates from North America were 30% and from Australia 17%, in European countries it varied from 5 to 50% with the highest rate found in Finland, whilst in Africa it is higher [16^19]. The resistance rate could vary among patients in the same city but with di erent countries of birth [20]. Some authors reported a higher resistance rate among H. pylori isolated from women than from men [16,21] but these data have not been con rmed by other authors [22]. Moreover, some authors found an increase in metronidazole resistance rate with the years of use [22,23], and others found the highest resistance rate in isolates from patients without ulcer (80%) as compared with patients with ulcer (32%) [21] Thus, considering this diversity in metronidazole resistance, it is important to know the prevalence in each area and in each population. No o cial recommendations exist on the suscept- Table 1 Number and percentage of patients with eradication or not when they were infected by a strain with a metronidazole (MET) MIC of 92 mg l 31, 4^8 mg l 31 or v16 mg l 31. MET MIC Number of patients H. pylori eradicated H. pylori not eradicated MIC 92 mgl (88.3%) 5 (11.6%) MIC 4^8 mg l (50%) 3 (50%) MIC v16 mg l (50%) 4 (50%)

4 186 M. Löpez-Brea et al. / FEMS Immunology and Medical Microbiology 24 (1999) 183^188 ibility methods, or on the cut-o points used to classify a strain as susceptible or resistant. For other bacteria, a strain is usually categorised as resistant when the tolerated antimicrobial concentration is higher than the concentration achieved in vivo at the site of infection, or in practical terms, in the serum. This criterion is not useful for H. pylori, due to the fact that serum concentrations are quite di erent from gastric mucosa ones and the concentration of the compound in the mucosa is very di cult to determine. Thus, a resistant strain should be de ned as a strain for which the likelihood of eradication by the speci c treatment is very low. It is necessary to perform clinical trials where the e ciency of a given regimen is correlated with bacteriological data and accurate susceptibility testing. According to the data obtained in this study, at least for children, a strain with an MIC v4 mgl 31 should be considered resistant because a decrease in the eradication rate is observed. However, most authors consider a strain resistant when the MIC is s 8mg l 31 or v8 mgl 31. Infection with caga-positive strains has been considered a marker of treatment failure by some authors [24] but not by others [25]. Some reports suggest that it is more di cult to eradicate the caganegative H. pylori strains than the caga-positive ones [11]. In our study no di erences in eradication rate were found among caga+- and caga3-infected patients, or in patients infected with strains that harbour the vaca s1 or vaca s2 alleles. Several studies have reported that when a patient is infected by a metronidazole-resistant strain, the possibility of obtaining eradication with treatment including metronidazole is lower than in patients infected with metronidazole-susceptible strains [15,26]. A 70 and 30% eradication rate has been found by Moayadi et al. [27], 100 and 53% by Ching et al. [21], 81 and 16% by DeCross [28] on susceptible and resistant H. pylori-infected patients. Other authors found 100 versus 82% in susceptible or resistant strains-infected patients [29] or eradication in eight out of nine versus 63 of 67 in metronidazole-resistant or -susceptible-infected patients [6], when triple therapy including metronidazole and clarithromycin was used. In 1997 Megraud reported a 96%, 78% and 45% eradication rate with a triple therapy including amoxicillin, metronidazole and lansoprazole, when the strains had a MIC to metronidazole of 6 2mg l 31, 2^8 mg l 31 or s 8 mg l 31, respectively [15]. Matsumoto et al. studied the eradication problem in a mouse model and found that when metronidazole-resistant infected mice were treated with 10, 32 or 100 mg kg 31 metronidazole, eradication was achieved in 0/5, 0/5 and 3/5 whilst the same gures for metronidazole-susceptible infected mice were 2/5 in the 10 mg kg 31 group, 4/5 in the 32 mg kg 31 group and 5/5 in the 100 mg kg 31 group [30]. Van der Hulst et al. reported that the addition of an acidsuppressing agent, such as a proton pump inhibitor, to the bismuth triple regimen may improve the usually impaired e cacy of this therapy in patients infected with metronidazole-resistant H. pylori, as they obtained an eradication rate of 82% [7]. Earlier data of these kinds of studies were obtained from patients su ering from peptic ulcer, and in these subjects the compliance is assumed to be excellent, due to the severity of the symptomatology. However, studies including patients with dyspepsia have been performed with similar results [7,25]. In conclusion, H. pylori eradication depends on the susceptibility of the strain to metronidazole, being higher in patients infected with susceptible H. pylori. However, when the MIC to metronidazole is v4 mg l 31 the eradication rate decreases to 50%. According to our data the caga or vaca status was not an important factor in treatment failure in the eradication of H. pylori. It is important to determine the susceptibility of the strains prior to treatment, and if this is not possible, at least to know the metronidazole resistance rate in each population to help in the selection of the anti-h. pylori therapy. Acknowledgements This work was supported by the Fondo de Investigaciones Sanitarias de la Seguridad Social FIS 95/ We wish to thank Brenda Ashley for her English language assistance.

5 M. Löpez-Brea et al. / FEMS Immunology and Medical Microbiology 24 (1999) 183^ References [1] Dunn, B.E., Cohen, H. and Blaser, M. (1997) Helicobacter pylori. Clin. Microbiol. Rev. 10, 720^741. [2] Patel, P., Mendall, M., Khulusi, S., North eld, T.C. and Strachan, D.P. (1994) Helicobacter pylori infection in childhood: risk factors and e ect on growth. Br. Med. J. 309, 1119^1123. [3] Graham, D.Y., Lew, G.M., Malaty, H.M., Evans, D.G., Evans, D.J. and Klein, P.D. et al. (1992) Factors in uencing the eradication of Helicobacter pylori with triple therapy. Gastroenterology 102, 493^496. [4] Bell, G.D., Powell, K., Burridge, S.M., Pallecaros, A., Jones, P.H. and Gant, P.W. et al. (1992) Experience with triple anti- Helicobacter pylori eradication therapy: side e ects and the importance of testing the pre-treatment bacterial isolate for metronidazole resistance. Aliment. Pharmacol. Ther. 6, 427^ 435. [5] Xia, H.X., Daw, M.A., Sant, S., Beattie, S., Keane, C.T. and O'Morain, C.A. (1993) Clinical e cacy of triple therapy in Helicobacter pylori associated duodenal ulcer. Eur. J. Gastroenterol. Hepatol. 5, 141^144. [6] Adamek, R.J., Suerbaum, S., Pfa enbach, B. and Opferkuch, W. (1998) Primary and acquired Helicobacter pylori resistance to clarithromycin, metronidazole and amoxicillin ^ in uence on treatment outcome. Am. J. Gastroenterol. 93, 386^389. [7] Van der Hulst, R.W.M., van der Ende, A., Homan, A., Roorda, P., Dankert, J. and Tytgat, G.N.J. (1998) In uence of metronidazole resistance on e cacy of quadruple therapy for Helicobacter pylori eradication. Gut 42, 166^169. [8] Lanbert, J. and Cardaci, G. (1995) Gastric mucosal pharmacology. A rational approach to Helicobacter pylori eradication. J. Gastroenterol. Hepatol. 10, 106^107. [9] Atherton, J. and Covacci, A. (1997) Pathogenic properties of Helicobacter pylori. Curr. Opin. Gastroenterol. 13, (Suppl. 1) 20^24. [10] Graham, D.Y., Genta, R.M., Graham, D.P. and Crabtree, J.E. (1996) Serum caga antibodies in asymptomatic subjects and patients with peptic ulcer ^ lack of correlation of IgG antibody in patients with peptic ulcer or asymptomatic Helicobacter pylori gastritis. J. Clin. Pathol. 49, 829^832. [11] Broute, N., Marais, A., Liu, D., Lamouliatte, H., Samoyeau, R. and Megraud, F. (1998) Cag negative status: a risk factor for failure of Helicobacter pylori triple therapies in non-ulcer dyspepsia. Gut 43, (Suppl. 2) A79. [12] Löpez-Brea, M., Alarcön, T. and Megraud, F. (1997) Diagnosis of Helicobacter pylori infection. Curr. Opin. Gastroenterol. 13, (Suppl. 1) 13^19. [13] Covacci, A., Censini, S. and Bugnoli, M. et al. (1993) Molecular characterization of the 128-kDa immunodominant antigen of Helicobacter pylori associated with cytotoxicity and duodenal ulcer. Proc. Natl. Acad. Sci. USA 90, 5791^5795. [14] Atherton, J.C., Cao, P., Richard, M.P.J., Tummuru, M.K.R., Blaser, M. and Cover, T.L. (1995) Mosaicism in vacuolating cytotoxin alleles of Helicobacter pylori. J. Biol. Chem. 270, 17771^ [15] Megraud, F. (1997) Resistance of Helicobacter pylori to antibiotics. Aliment. Pharmacol. Ther. 11, (Suppl. 1) 43^53. [16] Glupczynski, Y. (1992) European Multicenter Study Group on antibiotic susceptibility of Helicobacter pylori. Results of a mulicentre European survey in 1991 of metronidazole resistance in Helicobacter pylori. Eur. J. Clin. Microbiol. Infect. Dis. 11, 777^781. [17] Noach, L.A., Langenberg, W.L., Bertola, M.A., Dankert, J. and Tytgat, G.N. (1994) Impact of metronidazole resistance on eradication of Helicobacter pylori. Scand. J. Infect. Dis. 26, 321^326. [18] Westblom, T.U. and Unge, P. (1992) Drug resistance of Helicobacter pylori: memorandum from a meeting at the Sixth International Workshop on Campylobacter, Helicobacter and related organisms. J. Infect. Dis. 165, 974^975. [19] Vasquez, A., Valdez, Y., Gilman, R.H., McDonald, J.J., Westblom, T.U., Berg, D., Mayta, H. and Gutierrez, V. (1996) Metronidazole and clarithromycin resistance in Helicobacter pylori determined by measuring MICs of antimicrobial agents in colour indicator egg yolk agar in a miniwell format. J. Clin. Microbiol. 34, 1232^1234. [20] Banarvala, N., Davies, G.R., Adbi, Y., Claments, L., Rampton, D.S., Hardie, J.M. and Feldman, R.A. (1994) High prevalence of Helicobacter pylori metronidazole resistance in migrants to east London: relation with previous nitroimidazole exposure and gastroduodenal disease. Gut 35, 1562^1566. [21] Ching, C.K., Leung, K.P., Yung, R.W., Lam, S.K., Wong, B.C., Lai, K.C. and Lai, C.L. (1996) Prevalence of metronidazole resistant Helicobacter pylori strains among Chinese peptic ulcer disease patients and normal controls in Hong Kong. Gut 38, 675^678. [22] Löpez-Brea, M., Domingo, D., Sänchez, I. and Alarcön, T. (1997) Evolution of resistance to metronidazole and clarithromycin in Helicobacter pylori clinical isolated from Spain. J. Antimicrob. Chemother. 40, 279^281. [23] Van der Wouden, E.J., van Zwet, A.A., Vosmaer, G.D.C., Oom, J.A.J., de Jong, A. and Kleibeuker, J.H. (1997) Rapid increase in the prevalence of metronidazole resistant Helicobacter pylori in The Netherlands. Emerg. Infect. Dis. 3, 385^ 389. [24] Oderda, G., Ravanini, P., Malosso, M.C., Altare, F., Forni, M., Cerro, P., Caristo, P. and Fortina, G. (1996) Detection and quanti cation of serum antibodies to H. pylori speci c antigens of childhood HP gastritis. Gut 39, (Suppl. A) 51. [25] Raymond, J., Kalach, N., Bergeret, M., Benhamou, P.H., Barbet, J.P., Gendrel, D. and Dupont, C. (1998) E ect of metronidazole resistance on bacterial eradication of Helicobacter pylori in infected children. Antimicrob. Agents Chemother. 42, 1334^1335. [26] Lerang, F., Moum, B., Haug, J.B., Tolas, P., Breder, O., Aubert, E., Hoie, O. and Soberg, T. et al. (1997) Highly effective twice-daily triple therapies for Helicobacter pylori infection and peptic ulcer disease does in vitro metronidazole resistance have any clinical relevance? Am. J. Gastroenterol. 92, 248^253. [27] Moayyedi, P., Chalmers, D.M. and Axon, A.T. (1997) Patient factors that predict failure of omeprazole, clarithromycin and tinidazole to eradicate Helicobacter pylori. J. Gastroenterol. 32, 24^27.

6 188 M. Löpez-Brea et al. / FEMS Immunology and Medical Microbiology 24 (1999) 183^188 [28] De Cross, A.J., Marshall, B.J., McCallum, R.W., Ho man, S.R., Barrett, L.J. and Guerrant, R.L. (1993) Metronidazole susceptibility testing for Helicobacter pylori comparison of disk, broth and agar dilution methods and their clinical relevance. J. Clin. Microbiol. 31, 1971^1974. [29] Peitz, U., Nusch, A., Tillenburg, B., Stolte, M., Bo«rsch, G. and Labenz, J. (1996) High cure of H. pylori infection by one week therapy with omeprazole (ome), metronidazole (met) and clarithromycin (cla) despite a negative impact by met resistance. Gut 39, (Suppl. 2) A5. [30] Matsumoto, S., Washizuka, Y., Matsumoto, Y., Tawara, S., Ikeda, F., Yokota, Y. and Karita, M. (1997) Appearance of a metronidazole-resistant Helicobacter pylori strain in an infected-icr-mouse model and di erence in eradication of metronidazole resistant and sensitive strains. Antimicrob. Agents Chemother. 41, 2602^2605.