National and Regional US Antibiotic Resistance to Helicobacter Pylori: Lessons from a Clinical Trial.

Although antimicrobial resistance to Helicobacter pylori has increased globally, resistance patterns have been shown to differ in relation to sex, age, ethnicity, lifestyle habits, socioeconomic status, geographic distribution, and particularly with national antibiotic consumption rates.1Savoldi A. et al.Gastroenterology. 2018; 155: 1372-1382Abstract Full Text Full Text PDF PubMed Scopus (408) Google Scholar H pylori is not yet included in local, regional, and national susceptibility data. A 2015 study from the Houston VA Medical Center reported an increase of clarithromycin resistance among H pylori from 9.1% in 2009–2010 to 24.2% in 2011–2013; 20.3% of the isolates were resistant to metronidazole. H pylori resistance to tetracycline, amoxicillin, and rifabutin have remained historically low to absent in the United States.2Shiota S. et al.Clin Gastroenterol Hepatol. 2015; 13: 1616-1624Abstract Full Text Full Text PDF PubMed Scopus (111) Google Scholar Here we report antimicrobial susceptibility from a randomized, double-blind, placebo-controlled H pylori eradication study conducted in the United States from 2017 to 2018 from clinical study sites across 20 states. The clinical results have previously been reported.3Graham D.Y. et al.Ann Intern Med. 2020; 172: 795-802Crossref PubMed Scopus (34) Google Scholar These data represent the largest survey of H pylori antimicrobial susceptibility in more than 15 years. The study protocol and all amendments were approved by the respective institutional review boards of participating institutions. All patients provided written informed consent before participating in the study. Confidence intervals (CIs) for proportions and Fisher’s exact and Wilcoxon rank sum tests were applied using STATA 11 (College Station, TX). Analyses were 2-tailed, and a P < .05 was considered statistically significant. Four hundred fifty-five patients were enrolled in a double-blinded phase III clinical study that evaluated a fixed-dose triple-therapy combination of amoxicillin–rifabutin–omeprazole vs a fixed-dose dual-therapy comparator (amoxicillin–omeprazole). Minimum inhibitory concentrations (MICs) were obtained for 345 clinical isolates recovered from treatment-naive patients at 51 centers (20 states) across the United States, representing 76% of the total number of patients enrolled in the study (Supplementary Methods). The most common reason for failure of culture was improper storage or shipment of specimens. Additionally, a few isolates grew poorly on the Mueller-Hinton agar used for MIC determination, such that a reliable result could not be obtained. Of 345 patients, 131 were men (38%), and mean patient age was 46.4 ± 13 years (range, 18–70 years). Most patients (263, 76.2%) were white, 71 (20.6%) were black, and 11 (3.2%) were of other backgrounds (Asians were excluded from participation); overall, 208 (60.3%) identified as Hispanic. Enrollment was unevenly distributed across the different states. Across all sites, using the breakpoints listed above, resistance rates were 0.0% for rifabutin, 6.4% for amoxicillin (95% CI, 3.8%–9.0%), 17.4% for clarithromycin (95% CI, 13.6%–21.4%), and 43.6% for metronidazole (95% CI, 38.3%–48.9%). Dual resistance to metronidazole and clarithromycin was 10.5%. The MIC90 was 0.008 μg/mL for rifabutin, 0.125 μg/mL for amoxicillin, 8 μg/mL for clarithromycin, and 64 μg/mL for metronidazole. Most isolates (93.6%) tested were susceptible to the drug combination used in the trial (amoxicillin plus rifabutin), using European Committee on Antimicrobial Susceptibility Testing breakpoints. Metronidazole resistance was associated with an Hispanic background (48.3% vs 36.5% for non-Hispanics, P = .035) and with women (52.1% vs 29.8% for men, P < .0001) but not with age. Patients with clarithromycin resistance were slightly older (50.1 ± 11 years vs 45.6 ± 13 years, P < .01). Among the 71 isolates tested for levofloxacin and tetracycline resistance, 41 (57.8%) were resistant to levofloxacin and 2 (2.8%) were resistant to tetracycline. As observed with clarithromycin resistance, patients with levofloxacin resistance were older (53.3 ± 9 years vs 45.3 ± 11 years, P = .002). Importantly, almost one-fourth of patients (17/71 or 23.9%) with clarithromycin resistance had dual resistant to levofloxacin. Twenty-two of 71 (31%) isolates with clarithromycin resistance had dual resistance to metronidazole and levofloxacin; 16.9% of the isolates (12/71) were resistant to all 3 antibiotics. To better describe the data from a geographic perspective, the United States was divided into 3 regions, East, Central, and West, to allow visualization and comparison across regions (Table 1). Clarithromycin resistance ranged from 11.1% in the Western United States to 23.2% in the Eastern states (P = .03). Metronidazole resistance was high in all regions, ranging from 35.5% to 49.7%. Amoxicillin resistance ranged from 5.3% to 8.3%. Because of the smaller numbers tested for resistance to levofloxacin or tetracycline, it was not possible to detect whether there were significant regional differences.Table 1Geographic Distribution of Antimicrobial Resistance Patterns per US Region, 2017–2018AntibioticAll Isolates (N = 345)West (n = 108)Central (n = 86)East (n = 151)PAmoxicillin6.4 (22)8.3 (9)5.8 (5)5.3 (8).6Clarithromycin17.4 (60)11.1 (12)15.1 (13)23.2 (35).03Metronidazole43.6 (150)aMetronidazole susceptibilities were available for 107 isolates from the West region and from 344 isolates in total.35.5 (38)aMetronidazole susceptibilities were available for 107 isolates from the West region and from 344 isolates in total.43.0 (37)49.7 (75).08Rifabutin0.0 (0)0.0 (0)0.0 (0)0.0 (0)NDLimited Analysis (n = 71)(n = 19)(n = 19)(n = 33)Tetracycline2.8 (2)5.3 (1)5.3 (1)0.0 (0).3Levofloxacin57.8 (41)57.8 (11)68.4 (13)17 (51.5).5NOTE. Values are % (n). ND, not done.a Metronidazole susceptibilities were available for 107 isolates from the West region and from 344 isolates in total. Open table in a new tab NOTE. Values are % (n). ND, not done. Overall, we found that the proportion of US isolates with resistance to clarithromycin (17.6%), metronidazole (43.6%), and levofloxacin (57.8%) was higher compared with previous reports. As noted above, this study showed high rates of metronidazole resistance across all regions. No isolate was resistant to rifabutin. Amoxicillin and tetracycline resistance were both observed, albeit at low rates (amoxicillin resistance, 6.4%; tetracycline resistance, 2.8%). In both treatment arms of the prior study, amoxicillin resistance had a small but measurable effect on treatment outcome.3Graham D.Y. et al.Ann Intern Med. 2020; 172: 795-802Crossref PubMed Scopus (34) Google Scholar In contrast, resistance to clarithromycin, metronidazole, or levofloxacin is highly correlated with treatment failure when they are used in triple therapies. There was also a poor correlation between outcome (cure or failure) with individual isolates and the measured MIC resistance breakpoint for amoxicillin with the dual high-dose PPI and amoxicillin treatment arm. Interpretation of the effect of amoxicillin resistance on outcome in the dual-therapy regime was further complicated by the fact that the cure rate with dual therapy was low (ie, 57.7%), which was however consistent with prior US studies of similar therapy.4Laine L. et al.Aliment. Pharmacol Ther. 1998; 12: 377-382Crossref PubMed Scopus (9) Google Scholar, 5Attumi T.A. et al.Helicobacter. 2014; 19: 319-322Crossref PubMed Scopus (31) Google Scholar, 6Graham D.Y. et al.J Gastroenterol. 2010; 45: 816-820Crossref PubMed Scopus (39) Google Scholar Clarithromycin resistance differed significantly across the regions (P = .03), with the greatest resistance observed in the East. Metronidazole resistance was high in all regions. These results are consistent with the global increase in antimicrobial resistance and underscores the most recent H pylori guidelines to no longer prescribe clarithromycin, metronidazole, or levofloxacin–containing triple therapies.7Liou J.M. et al.Gut. 2020; 69: 2093-2112Crossref PubMed Scopus (76) Google Scholar,8El-Serag H.B. et al.Clin Gastroenterol Hepatol. 2018; 16: 992-1002Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar The general unavailability of readily accessible H pylori culture and susceptibility testing requires treating physicians to pay close attention to the results of testing for cure to alert them when therapies become ineffective.9Graham D.Y. Antibiotics (Basel). 2020; 9: 671Crossref Scopus (24) Google Scholar The recently introduced rifabutin triple therapy (Talicia) and bismuth quadruple therapy are the only currently available regimens in the United States for which H pylori resistance is rare. The commercially available bismuth formulation, Pylera, is often inconvenient to prescribe because it is packaged for 10-day therapy, whereas treatment guidelines recommend 14-day therapy for optimal effectiveness.9Graham D.Y. Antibiotics (Basel). 2020; 9: 671Crossref Scopus (24) Google Scholar Local antimicrobial resistance rates are needed to allow clinicians to reliably and successfully eradicate H pylori infections. Kristina G. Hulten, PhD (Conceptualization: Equal; Data curation: Equal; Formal analysis: Equal; Funding acquisition: Equal; Investigation: Equal; Methodology: Equal; Project administration: Equal; Resources: Equal; Supervision: Equal; Validation: Equal; Writing – original draft: Equal; Writing – review & editing: Equal). Linda B. Lamberth, BA (Data curation: Supporting; Investigation: Supporting; Methodology: Supporting; Writing – original draft: Supporting). Ira N. Kalfus, MD (Conceptualization: Equal; Formal analysis: Equal; Investigation: Equal; Methodology: Equal; Project administration: Supporting; Resources: Supporting; Supervision: Supporting; Validation: Equal; Writing – review & editing: Equal). David Y. Graham, MD (Conceptualization: Lead; Data curation: Equal; Formal analysis: Equal; Funding acquisition: Supporting; Investigation: Lead; Methodology: Equal; Project administration: Equal; Resources: Equal; Supervision: Lead; Validation: Equal; Writing – original draft: Lead; Writing – review & editing: Lead). H pylori was isolated from gastric biopsies obtained as part of a screening in a nationwide US H pylori eradication study. One antral and 1 corpus biopsy were combined into 1 container containing transport media consisting of cysteine media (7 g/L Difco casamino acids, 7 g/L Difco peptone, 1.4 g/L Difco yeast extract, 3.5 g/L NaCl, 0.14 g/L L-cysteine, 0.7 g/L dextrose pH 7.0) with 20% glycerol. The samples were frozen on dry ice and shipped to the Dr Edward O. Mason Infectious Disease Research Laboratory at Texas Children’s Hospital where they were stored at –70°C until cultured. Biopsies were processed under sterile conditions, and the mechanically disrupted biopsy tissue material was cultured on selective (10 μg/mL nalidixic acid, 5 μg/mL trimethoprim, 3 μg/mL vancomycin, and 2 μg/mL amphotericin B; Sigma-Aldrich) and nonselective brain heart infusion (BD BBL) agar plates with 7% horse blood (Cocalico Biologicals, Stevens, PA) at 37°C under microaerophilic conditions obtained using the Anoxomat system (Advanced Instruments, Inc). Agar plates were incubated for up to 7 days. Most isolates grew in 3–4 days. H pylori was identified from the agar plates, and pure cultures were verified by oxidase, catalase, and urease biochemical tests (BD BBL) and Gram stain. All isolates were tested for MICs for amoxicillin, clarithromycin, metronidazole, and rifabutin as part of the treatment study by agar dilution using Clinical and Laboratory Standards Institute (CLSI) methods and guidelines.1Wayne P.A. Clinical and Laboratory Standards InstituteClinical and Laboratory Standards Institute.2018Google Scholar After completion of the treatment study, a subset of isolates were reisolated from frozen stocks and evaluated for tetracycline and levofloxacin resistance using the same methods. These isolates were not selected based on any specific characteristics other than their availability and viability. All antibiotics were obtained from US Pharmacopeia (North Bethesda, MD). H pylori ATCC 43504 was included as a control. Because CLSI breakpoints have been determined only for clarithromycin, CLSI and the European Committee on Antimicrobial Susceptibility Testing (EUCAST, www.eucast.org) interpretive breakpoints were used to determine resistance rates.1Wayne P.A. Clinical and Laboratory Standards InstituteClinical and Laboratory Standards Institute.2018Google Scholar,2Wayne P.A. Clinical and Laboratory Standards InstituteClinical and Laboratory Standards Institute.2018Google Scholar Resistance breakpoints were >0.125 μg/mL for amoxicillin (EUCAST), ≥1 μg/mL for clarithromycin (CLSI, EUCAST), >8 μg/mL for metronidazole (EUCAST), and >1 μg/mL for levofloxacin, rifabutin, and tetracycline (EUCAST).1Wayne P.A. Clinical and Laboratory Standards InstituteClinical and Laboratory Standards Institute.2018Google Scholar,2Wayne P.A. Clinical and Laboratory Standards InstituteClinical and Laboratory Standards Institute.2018Google Scholar