When organic causes for upper abdominal symptoms, or dyspepsia, are suspected in children, esophagogastroduodenoscopy (EGD) should be undertaken, including biopsies for rapid urease testing and histology; culture for Helicobacter pylori may also be performed (1–3). EGD reliably identifies H pylori–associated peptic ulcer disease, one of the few undisputed indications for H pylori eradication treatment in children. Although these issues are controversial, there is some evidence that children who are positive for H pylori experience dyspeptic symptoms significantly more often than do children who are negative and benefit from eradication of the infection (4–10). Culturing H pylori may be particularly important for patients living in regions with a high prevalence of clarithromycin-resistant strains (11,12). Clarithromycin is widely used as part of the first-line combination therapies because it is well tolerated and effective against susceptible H pylori; however, in the case of resistance, treatment regimens including clarithromycin have a high failure rate (13,14).
In Vienna, the prevalence of clarithromycin resistance among isolates from the pediatric population is increasing (4,15). Thus, H pylori eradication therapy tailored to antimicrobial susceptibility testing is recommended and the “test and treat” strategy based on noninvasive tests is discouraged (1–3). Pediatric endoscopy units are still relatively rare and often cannot accommodate all of the children who need EGD. In addition, patients and their parents as well as many pediatricians may be opposed to EGD and the associated anesthesia.
Recently, a stool polymerase chain reaction (PCR) was established (16). Stool PCR not only tests for H pylori but it is also the only noninvasive certified in vitro diagnostic method that provides information on the clarithromycin susceptibility of H pylori. Both of these properties have satisfactory sensitivity and excellent specificity in dyspeptic children (16).
Since June 2007, our approach to dyspepsia comprised noninvasive tests for H pylori including stool PCR in children in whom EGD was not needed or was rejected. When stool PCR results revealed clarithromycin sensitivity, clarithromycin-containing eradication treatment was prescribed. EGD was recommended in the case of warning signs/symptoms or clarithromycin resistance shown by PCR. It has not been evaluated whether such an approach places those children receiving eradication treatment based solely on stool PCR results at a disadvantage. In the present retrospective cohort study, we tested whether a treatment concept based on noninvasive diagnostic tools, shown in our previous study (16) to be less reliable in detecting resistance to clarithromycin than the invasive criterion standard, affects treatment success. We determined the eradication rate in H pylori–infected children whose therapy is tailored to stool PCR (stool PCR group) compared with that in children whose therapy is based on culture followed by antibiogram (gastric biopsy group). In addition, we examined the acquisition of resistance to clarithromycin after an unsuccessful initial treatment course in both groups.
PATIENTS AND METHODS
Human Subjects Protection
The ethics committee of the St Anna Children's Hospital, Vienna, reviewed and approved the present study.
Study Design and Setting
This was a retrospective cohort study to assess the eradication rate of H pylori and the prevalence of acquired clarithromycin resistance among children after an initial course of tailored triple eradication therapy. For this assessment we collected data on 96 consecutive children who for the first time had been prescribed treatment to eradicate H pylori between June 2007 and September 2009, at the pediatric gastroenterology outpatient clinic of the St Anna Children's Hospital. Previously treated children having failed 1 or more courses of therapy were excluded. Patients were identified from hospital administrative databases.
Assigning the Children to Either the Gastric Biopsy Group or the PCR Group
For the diagnosis of H pylori infection in children undergoing EGD, we routinely performed the rapid urease test, histology, and culture (4); minimal diagnostic criteria for an infection were positive results of both rapid urease test and histology or a positive culture result alone. In children who did not undergo EGD, positive results of both a noninvasive standard test and stool PCR defined the presence of H pylori. The noninvasive standard tests were the 13C-urea breath test or a monoclonal stool enzyme immunoassay, the latter mainly in children younger than 5 years old. Whenever EGD was not necessary or possible, indications for noninvasive H pylori testing were epigastric pain or discomfort associated with fullness and early satiety, nausea, iron deficiency, and follow-up after H pylori eradication therapy. Criteria suggesting the need for EGD were warning signs or symptoms such as blood loss, weight loss, recurrent vomiting, dysphagia, nocturnal pain, anemia, malabsorption, and concomitant disease with possible involvement of the gastrointestinal tract. In addition, when family history was positive for peptic ulcer disease or gastric malignancies or when stool PCR revealed H pylori resistant to clarithromycin, we strongly suggested EGD. Whenever only noninvasive H pylori tests were performed, eradication treatment was tailored according to the results of stool PCR (stool PCR group); in the case of EGD, treatment was tailored according to antibiogram (gastric biopsy group).
Children were reevaluated not earlier than 5 weeks after the end of the eradication treatment by a noninvasive standard test and stool PCR, also allowing for the determination of the posttreatment clarithromycin susceptibility status. Any documentation in the medical records of new symptoms or the aggravation of preexisting complaints reported by the patients and/or their parents as being associated with treatment was classified as a possible adverse effect.
EGD was performed under general anesthesia using a pediatric gastroscope (GIF-Q165, Olympus, Hamburg, Germany). Routinely 4 biopsies were taken from the gastric antrum (2 for histology, 1 for rapid urease test, and 1 for culture) and 3 from the gastric body (2 for histology, and 1 for culture).
Culture and Epsilometer Test (E-test)
After being transported in Portagerm pylori (Biomerieux, Marcy L'Etoile, France) biopsy samples were homogenized and culture followed by antibiogram was performed as reported previously (4). Breakpoints of resistance to clarithromycin, metronidazole, amoxicillin, levofloxacin, tetracycline, and rifampin were minimal inhibitory concentration values of ≥1, 16, 4, 2, 2, and 4 μg/mL, respectively.
For PCR, stool specimens had to be fresh, from the same morning or the day before. Refrigeration was recommended until transportation. Samples arrived at the laboratory on the same day. They were divided into aliquots and stored at −20°C until analysis. Using 200 mg stool, DNA was extracted with a QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Stool DNA extracts were analyzed by fluorescence resonance energy transfer real-time PCR and melting curve analysis (H pylori ClariRes assay, Ingenetix, Vienna, Austria) in combination with LightCycler-FastStart DNA Master SYBR Green I (Roche Molecular Biochemicals, Mannheim, Germany). A DNA probe labeled with the fluorophore Cy5 includes the sites of the mutations on the H pylori 23S rRNA gene responsible for resistance to clarithromycin and has 100% homology to the sensitive wild-type genotype. Analysis was performed according to the manufacturer's instructions in a LightCycler apparatus (Roche Diagnostics, Mannheim, Germany). Samples were run in duplicate. Results were considered positive for H pylori if at least 1 of the reactions showed a specific melting curve. A melting temperature of 63°C corresponds to the clarithromycin-sensitive wild-type genotype; in resistant genotypes the melting temperature is 54°C for mutations A2142G and A2143G and 58°C for mutation A2142C. In H pylori–negative samples, the presence of a melting peak at 47°C (internal control) indicates PCR conditions that allow amplification.
Detection of H pylori Antigen in Feces
H pylori antigen in stool specimens was determined by Amplified IDEIA Hp StAR (DakoCytomation, Cambridgeshire, UK), a sandwich-type amplified enzyme immunoassay using monoclonal antibodies against H pylori antigens. The test was performed according to the manufacturer's instructions. Spectrophotometer results with absorbance values (A450/630) of ≥0.150 were considered positive.
13C-urea Breath Test
Breath samples were collected before and 15 minutes after ingestion of 200 mL of orange juice containing the 13C-urea reagent (13C test set with 75 mg of urea; Zamponi-Diagnostik, Judenburg, Austria). Samples were measured using an IR300-spectrophotometer. A test result was positive if the delta over base was ≥3.5/1000 (17,18).
Children received triple therapy consisting of a proton pump inhibitor, amoxicillin, and either clarithromycin, metronidazole, or levofloxacin. Treatment duration was 7 days except in patients younger than 12 years of age infected with H pylori that was resistant to both clarithromycin and metronidazole. These children received a 14-day triple therapy containing metronidazole. Adolescents with H pylori revealing such a dual resistance were given a 7-day levofloxacin-containing regimen. All of the patients were instructed to take omeprazole, esomeprazole, or pantoprazole 15 minutes before breakfast and dinner at a dose of 0.8 to 2 mg kg−1 day−1. Antibiotics were recommended after meals. Doses of antibiotics prescribed per day were as follows: 40 to 60 mg/kg of amoxicillin, 15 to 25 mg/kg of clarithromycin, 20 to 30 mg/kg of metronidazole, and 15 to 20 mg/kg of levofloxacin. Patients received a “medication calendar” with the names of the prescribed drugs, doses, and time of intake.
Statistical analyses were performed using independent-samples t test and 2-tailed Fisher exact test, with P < 0.05 considered significant (SPSS version 15, SPSS Inc, Chicago, IL). The adjusted Wald method was used to calculate 95% confidence intervals. Median, range, and interquartile range (IQR) were calculated for diagnosis to treatment intervals. The Mann-Whitney U test was performed to assess a difference between these intervals when comparing the gastric biopsy group with the stool PCR group.
Baseline characteristics of the 96 children enrolled in the present retrospective study (age, sex, ethnic origin) were not significantly different between the gastric biopsy and the stool PCR group (Table 1). Among these 96 children, 84 (87.5%) had an immigrant background.
Among all of the children who were prescribed eradication treatment (n = 96), 27 were endoscoped without prior stool PCR testing. In the remaining 69 children, stool PCR results were available: 14 showed resistance (20.3%) and 53 showed susceptibility to clarithromycin, whereas 2 were false-negative in the PCR. In the latter 2 cases, children underwent EGD to adhere to the concept of tailored treatment. Of the 14 children who tested clarithromycin resistant by stool PCR and were therefore advised to undergo EGD, only 2 (14.3%) declined. Thus, 41 children were assigned to the gastric biopsy group and 55 to the stool PCR group.
Among the 41 gastric biopsy group children, antral nodularity was detected in 35 (85.4%), whereas antral hyperemia without macroscopic nodularity was found in 3 (6.5%). Peptic ulcer disease was diagnosed in 4 of the 41 children (9.8%), 2 having gastric ulcer and 2 duodenal ulcer. Erosive bulbitis was detected in 1 child (2.4%) and esophagitis in 3 (7.3%). On histological examination, all of the gastric biopsies revealed chronic gastritis being classified as mild in 6 (14.6%), moderate in 17 (41.5%), and severe in 18 children (43.9%). Regarding detection of H pylori, in comparison with culture histology was false-negative in 1 child (2.4%) and rapid urease testing (RUT) in 4 other children (9.7%); all of the positive histology and/or RUT results were confirmed by culture.
Pretreatment Antimicrobial Susceptibility
Overall, infection with H pylori resistant to clarithromycin was diagnosed in 16 of the 96 children (16.7%) (Table 2) and was more prevalent in nonimmigrant children (3/12, 25%) than in those with an immigrant background (13/84, 15.5%); however, this difference was not statistically significant (P = 0.42).
Of the 41 gastric biopsy group children, 14 (34.1%) harbored H pylori resistant to clarithromycin. H pylori resistant to metronidazole was detected in 10 children (24.4%) (Fig. 1, Table 2). Five H pylori strains (12.2%) exhibited dual resistance to clarithromycin and metronidazole (Fig. 1). All of the H pylori strains were susceptible to amoxicillin, levofloxacin, tetracycline, and rifampin.
Of the 55 stool PCR group children, only 2 (3.6%) harbored H pylori resistant to clarithromycin (Fig. 1, Table 2). Because of our approach of advising patients with a clarithromycin-resistant infection to undergo EGD, clarithromycin resistance was significantly less common in the stool PCR group than in the gastric biopsy group (P < 0.001).
Clarithromycin was used more often and metronidazole less often in the stool PCR group than in the gastric biopsy group (P < 0.001 for each comparison) (Table 2). The use of levofloxacin did not differ in the 2 groups. One child each in the gastric biopsy and the stool PCR group received a metronidazole-containing anti–H pylori triple therapy despite clarithromycin susceptibility because of a concomitant infection with Giardia lamblia (Fig. 1).
The median time from diagnosis to treatment was 20.5 days (range 11–36, IQR 7.0) in the gastric biopsy group versus 14 days (range 2–142, IQR 18.0) in the PCR group (P = 0.099).
Overall, H pylori was eradicated in 73 of the 96 children (76.0%) as assessed by follow-up evaluation (mean time until follow-up 65 days, range 38–160 days). There was no significant difference between the gastric biopsy and stool PCR groups regarding eradication rate (Table 2). Infection was cleared in 59 of 78 (75.6%) children after clarithromycin-containing regimes versus 14 of 18 (77.8%) children, when treatments not containing clarithromycin were used (P = 1.0).
There was a trend toward a higher eradication rate in nonimmigrant children than in those with an immigrant background (100% vs 72.6%, P = 0.064). Comparing only those children with an immigrant background (n = 84), H pylori was eradicated in 28 of 39 children from the gastric biopsy group versus in 33 of 45 children from the stool PCR group (71.8% vs 73.3%, P = 1.0).
Posttreatment Resistance to Clarithromycin
Posttreatment stool PCR gave false-negative results in 4 of 23 cases (17.4%). No false-positive results were found. In comparison with the 16.7% clarithromycin resistance rate before eradication treatment, clarithromycin resistance was detected in 47.4% of children with positive PCR results after treatment (P = 0.006). Among the children who failed to respond to a clarithromycin-containing triple therapy (n = 19, Table 2), posttreatment stool PCR revealed clarithromycin susceptibility in 10, resistance in 5, and a false-negative result in 4.
Of the 11 (26.8%) gastric biopsy group children with persistent H pylori infection at follow-up, 8 had received clarithromycin-containing therapy for strains susceptible to clarithromycin. One of the 8 tested false-negative by posttreatment PCR, 4 still harbored H pylori sensitive to clarithromycin, whereas 3 children acquired secondary resistance to clarithromycin (Table 2). Among the 4 children with persisting clarithromycin susceptibility, 1 patient reported incomplete pill intake and another discontinued therapy because of nausea. Of the 30 successfully treated gastric biopsy group patients, 1 reported headache but still completed the therapy.
Of the 12 (21.8%) stool PCR group children with persistent H pylori infection, 11 had been harboring clarithromycin-sensitive strains before treatment. After treatment, 3 of the 11 tested falsely negative by PCR, 6 were still infected with strains susceptible to clarithromycin, whereas 2 children acquired secondary resistance to clarithromycin (Table 2). Among all of the stool PCR group patients, 2 developed an amoxicillin-induced exanthema and 1 complained of vomiting once during the treatment course. All 3 patients still completed treatment and cleared the H pylori infection.
Thus, clarithromycin resistance was acquired by 4.1% (2/49) of strains in the stool PCR group versus 12% (3/25) in the gastric biopsy group (P = 0.33). The secondary clarithromycin resistance rate was 33.3% (3/9) in the former and 60% (6/10) in the latter group (P = 0.37, Table 2).
Test Characteristics of Stool PCR
Compared with the noninvasive standard tests, the sensitivity of pretreatment PCR was shown to be 97.1% (67/69); in the 2 false-negative cases, EGD confirmed the infection with a clarithromycin-susceptible strain. An additional 12 children had been tested by stool PCR and the invasive criterion standard, both of which revealed a clarithromycin-resistant strain. Thus, concerning the detection of clarithromycin resistance by PCR, the positive predictive value was 100%. The test characteristics of posttreatment stool PCR exhibited a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 82.6%, 100%, 100%, 94.8%, and 95.8%, respectively.
In the present retrospective cohort study, we tested whether the eradication rate is lower in H pylori–infected children whose therapy is tailored according to stool PCR (stool PCR group) than in those treated with EGD followed by culture and antibiogram (gastric biopsy group). Our results indicate that the eradication rate is not adversely affected by the noninvasive approach of testing (Table 2). In addition, we examined the acquisition of resistance to clarithromycin after an unsuccessful initial treatment course. We found that the prevalence of H pylori strains showing resistance to clarithromycin significantly increased in children who failed an eradication therapy.
The major advantage of stool PCR is that it is noninvasive. Accordingly, the costs of stool PCR are considerably lower than those of EGD, followed by culture and antibiogram. This difference is even more pronounced when expenses for general anesthesia and hospital admission are taken into account. Therefore, the integration of stool PCR into the concept of tailored treatment could further increase the cost-effectiveness of this concept, which was already demonstrated as being economical in the case of conventional susceptibility testing–based therapy in adults (19). Additionally, the results of stool PCR can be available within hours, whereas for culture including antibiogram results, at least 6 days are required. This potential advantage of stool PCR could not be realized in the present study because the test was performed routinely only once per week and also because of the shortage of follow-up appointments in the routine of a busy outpatient clinic, thus delaying the prescription of treatment in both groups. The interval from diagnosis to treatment was shorter in the PCR group than in the culture group, even though this difference was not statistically significant. The most important advantage of culture followed by antibiogram is that it generates more extensive results by testing susceptibility to more antibiotics than stool PCR.
The importance of tailored treatment and strict compliance with therapy is highlighted by the 100% eradication rate that was achieved in nonimmigrant children. The excellent treatment success in the native PCR group children (n = 10) supports the usefulness of a noninvasive approach including the stool PCR. In contrast, in children with an immigrant background, the eradication rate was lower (72.6%), no matter whether their treatment had been tailored according to culture or PCR results (eradication in 71.8% vs 73.3%); however, the difference in terms of H pylori eradication in children without and with an immigrant background was statistically not significant. Nevertheless, the small proportion of native children included in the study does not allow us to draw reliable conclusions regarding a different treatment success in these 2 groups of children.
The rate of primary clarithromycin resistance was significantly higher in the gastric biopsy group than in the stool PCR group (Table 2). This difference is attributable to our patient care program. This program, to optimize therapy, proposes that children with H pylori resistant to clarithromycin (as shown by stool PCR) undergo EGD to gain information on resistance to additional antibiotics by culture followed by antibiogram. Consistently, clarithromycin-containing triple therapy was used significantly less often and metronidazole-containing therapy was used significantly more often in the gastric biopsy group than in the stool PCR group (Table 2). The poor eradication rate in the gastric biopsy group is unlikely to be because of the higher rate of clarithromycin resistance in this group. On the contrary, treatment success in the gastric biopsy group was marginally higher in children harboring H pylori resistant to clarithromycin than in those harboring susceptible strains (78.6% vs 70.3%, P = 0.71).
As shown by pretreatment stool PCR and culture and posttreatment stool PCR, secondary clarithromycin resistance emerges at a high rate following an unsuccessful initial eradication treatment (16.7% vs 47.4%, P = 0.006). The present finding concurs with data from various studies in adults and children. Most of these studies cross-sectionally compare 2 groups of patients, those never treated for H pylori infection versus previously treated ones (20–22). Only limited data are available from longitudinal assessments of the emergence of clarithromycin resistance in 1 group of pediatric patients undergoing antimicrobial susceptibility testing before treatment and as part of the follow-up diagnostics. In the context of such a pediatric study, Kalach et al reported 14 children failing an initial clarithromycin-containing triple therapy for an infection with clarithromycin-susceptible H pylori. Three of them were shown to be infected with a resistant strain on follow-up culture (23). These data are also in good agreement with ours, showing the emergence of secondary clarithromycin resistance in 5 of 15 patients infected with clarithromycin-susceptible H pylori before treatment.
Elsewhere, anti–H pylori triple therapy in children based on antibiogram achieved an eradication rate >90% (24). The eradication rate was similarly high in adults after administration of therapy based on fecal H pylori clarithromycin-susceptibility testing using molecular methods (25). In contrast, tailored treatment in our patients yielded an unexpectedly low eradication rate of only 76%; however, our data, which do not result from a randomized controlled trial (RCT) under optimized conditions, are still comparable with per-protocol treatment efficacy of some pediatric RCTs using similar drug combinations (26–28). Furthermore, it is well known that in clinical practice, the efficacies of treatments are usually lower than those obtained in RCTs (29). The data from the Pediatric European Register for Treatment of Helicobacter pylori revealed an eradication rate of only 64% after different triple therapies in children with an 18% prevalence of clarithromycin resistance (30). In our opinion, these data do reflect real life better than those obtained in the somewhat artificial setting of an RCT.
The comparison of the 2 methods, stool PCR versus endoscopy, biopsy, and culturing, is definitely affected by the retrospective cohort design of our study. Nevertheless, we believe that the present study is appropriate for delineating the role of stool PCR in the routine workup and guidance of treatment of H pylori infection in children. Moreover, for an RCT addressing our specific research questions, either all children should receive endoscopy or the study should be restricted to those without the need for EGD; however, considering the high prevalence of clarithromycin-resistant strains in our pediatric population, empirical treatment not including clarithromycin must be applied in a group of patients. Both scenarios have problematic implications and their feasibility may be doubtful from an ethical point of view.
In summary, the results of the present study show that in comparison with EGD followed by culture and antibiogram a noninvasive approach using stool PCR may be equally effective for tailoring eradication treatment for pediatric H pylori infection. Furthermore, stool PCR is a useful tool for the assessment of secondary clarithromycin resistance in children who failed an eradication treatment. It is conceivable that the introduction of stool PCR, also allowing for clarithromycin susceptibility testing as a further noninvasive routine test, may be helpful in controlling the emergence of multiresistant H pylori.
1. Gold BD, Colletti RB, Abbott M, et al. Helicobacter pylori infection in children: recommendations for diagnosis and treatment. J Pediatr Gastroenterol Nutr 2000; 31:490–497.
2. Drumm B, Koletzko S, Oderda G. Helicobacter pylori infection in children: a consensus statement. European Paediatric Task Force on Helicobacter pylori. J Pediatr Gastroenterol Nutr 2000; 30:207–213.
3. Jones N, Fallone C, Flook N, et al. Canadian Helicobacter Study Group Consensus Conference: update on the approach to Helicobacter pylori infection in children and adolescents—an evidence-based evaluation. Can J Gastroenterol 2005; 19:399–408.
4. Vécsei A, Kipet A, Innerhofer A, et al. Time trends of Helicobacter pylori resistance to antibiotics in children living in Vienna, Austria. Helicobacter 2010; 15:214–220.
5. Rutigliano V, Ierardi E, Francavilla R, et al. Helicobacter pylori and nonulcer dyspepsia in childhood: clinical pattern, diagnostic techniques, and bacterial strains. J Pediatr Gastroenterol Nutr 1999; 28:296–300.
6. Oderda G, Marinello D, Lerro P, et al. Dual vs. triple therapy for childhood Helicobacter pylori gastritis: a double-blind randomized multicentre trial. Helicobacter 2004; 9:293–301.
7. Farrell S, Milliken I, Murphy JL, et al. Nonulcer dyspepsia and Helicobacter pylori eradication in children. J Pediatr Surg 2005; 40:1547–1550.
8. Carelli AP, Patricio FR, Kawakami E. Carditis is related to Helicobacter pylori infection in dyspeptic children and adolescents. Dig Liver Dis 2007; 39:117–121.
9. Francavilla R, Lionetti E, Castellaneta SP, et al. Improved efficacy of 10-day sequential treatment for Helicobacter pylori eradication in children: a randomized trial. Gastroenterology 2005; 129:1414–1419.
10. Spee LA, Madderom MB, Pijpers M, et al. Association between Helicobacter pylori and gastrointestinal symptoms in children. Pediatrics 2010; 125:e651–e669.
11. Kalach N, Benhamou PH, Campeotto F, et al. Clarithromycin resistance and eradication of Helicobacter pylori in children. Antimicrob Agents Chemother 2001; 45:2134–2135.
12. Dore MP, Leandro G, Realdi G, et al. Effect of pretreatment antibiotic resistance to metronidazole and clarithromycin on outcome of Helicobacter pylori therapy: a meta-analytical approach. Dig Dis Sci 2000; 45:68–76.
13. Kato S, Konno M, Maisawa S, et al. Results of triple eradication therapy in Japanese children: a retrospective multicenter study. J Gastroenterol 2004; 39:838–843.
14. Elitsur Y, Lawrence Z, Russmann H, et al. Primary clarithromycin resistance to Helicobacter pylori and therapy failure in children: the experience in West Virginia. J Pediatr Gastroenterol Nutr 2006; 42:327–328.
15. Crone J, Granditsch G, Huber WD, et al. Helicobacter pylori in children and adolescents: increase of primary clarithromycin resistance, 1997–2000. J Pediatr Gastroenterol Nutr 2003; 36:368–371.
16. Vécsei A, Innerhofer A, Binder C, et al. Stool polymerase chain reaction for Helicobacter pylori detection and clarithromycin susceptibility testing in children. Clin Gastroenterol Hepatol 2010; 8:309–312.
17. Bazzoli F, Cecchini L, Corvaglia L, et al. Validation of the 13C-urea breath test for the diagnosis of Helicobacter pylori infection in children: a multicenter study. Am J Gastroenterol 2000; 95:646–650.
18. Kato S, Ozawa K, Konno M, et al. Diagnostic accuracy of the 13C-urea breath test for childhood Helicobacter pylori infection: a multicenter Japanese study. Am J Gastroenterol 2002; 97:1668–1673.
19. Romano M, Marmo R, Cuomo A, et al. Pretreatment antimicrobial susceptibility testing is cost saving in the eradication of Helicobacter pylori. Clin Gastroenterol Hepatol 2003; 1:273–278.
20. Romano M, Iovene MR, Russo MI, et al. Failure of first line eradication treatment significantly increases prevalence of antimicrobial-resistant Helicobacter pylori clinical isolates. J Clin Pathol 2008; 61:1112–5.
21. Raymond J, Lamarque D, Kalach N, et al. High level of antimicrobial resistance in French Helicobacter pylori isolates. Helicobacter 2010; 15:21–27.
22. Agudo S, Alarcon T, Cibrelus L, et al. High percentage of clarithromycin and metronidazole resistance in Helicobacter pylori clinical isolates obtained from Spanish children. Rev Esp Quimioter 2009; 22:88–92.
23. Kalach N, Benhamou PH, Dupont C, et al. Helicobacter pylori in children: acquisition of antimicrobial resistance after an initial course of treatment. J Clin Microbiol 2001; 39:3018–3019.
24. Arenz T, Antos D, Russmann H, et al. Esomeprazole-based 1-week triple therapy directed by susceptibility testing for eradication of Helicobacter pylori infection in children. J Pediatr Gastroenterol Nutr 2006; 43:180–184.
25. Kawai T, Yamagishi T, Yagi K, et al. Tailored eradication therapy based on fecal Helicobacter pylori clarithromycin sensitivities. J Gastroenterol Hepatol 2008; 23(suppl 2):S171–S174.
26. Gottrand F, Kalach N, Spyckerelle C, et al. Omeprazole combined with amoxicillin and clarithromycin in the eradication of Helicobacter pylori in children with gastritis: a prospective randomized double-blind trial. J Pediatr 2001; 139:664–668.
27. Faber J, Bar-Meir M, Rudensky B, et al. Treatment regimens for Helicobacter pylori infection in children: is in vitro susceptibility testing helpful? J Pediatr Gastroenterol Nutr 2005; 40:571–574.
28. Cadranel S, Bontemps P, Van Biervliet S, et al. Improvement of the eradication rate of Helicobacter pylori gastritis in children is by adjunction of omeprazole to a dual antibiotherapy. Acta Paediatr 2007; 96:82–86.
29. Fennerty MB, Lieberman DA, Vakil N, et al. Effectiveness of Helicobacter pylori therapies in a clinical practice setting. Arch Intern Med 1999; 159:1562–1566.
30. Oderda G, Shcherbakov P, Bontems P, et al. Results from the Pediatric European Register for Treatment of Helicobacter pylori (PERTH). Helicobacter 2007; 12:150–156.