ID Lectures / Resident and fellow education

Helicobacter pylori

Diagnosis, treatment selection, resistance, and confirmation of eradication.

Resident core · Optional fellow section · Clinical and laboratory reference · Edited November 10, 2026

This lecture is intended as a supplementary resource for clinicians and trainees to review evidence and discuss clinical cases.

Resident core

1. Indications for testing and treatment

Testing is most useful when a positive result would lead to treatment. Confirmed active H. pylori infection generally warrants eradication, even without symptoms. Serology cannot distinguish current from previous infection or document cure.

Clinical settingContext
Current or previous peptic ulcer diseaseIncluding patients using NSAIDs or aspirin. Eradication reduces ulcer recurrence; some patients still require gastroprotection.
Gastric MALT lymphoma, gastric atrophy or intestinal metaplasia, early gastric cancer resectionEradication is indicated but does not remove the need for appropriate GI surveillance or oncologic care.
Uninvestigated dyspepsiaNoninvasive test-and-treat is appropriate in selected adults younger than 60 without features that justify endoscopy; apply individualized cancer risk assessment.
Unexplained iron deficiency anemia or ITPConsider testing after appropriate evaluation for other causes.
Higher gastric cancer risk or adult household contact of an infected personACG 2024 expands testing to selected high-risk groups, including first-degree family history of gastric cancer and adult household contacts.

Routine universal screening of low-risk asymptomatic adults is not established. Alarm findings and elevated cancer risk may justify endoscopy rather than an isolated stool or breath test.

Deferral is reasonable during an acute competing illness, pregnancy when treatment can safely wait, or when severe interactions, intolerance, or inability to adhere make immediate eradication impractical. The reason for deferral and the opportunity for reassessment remain relevant to follow-up. Permanent observation of confirmed infection is generally not recommended.
Deferring treatment or discontinuing further eradication attempts

Confirmed active infection: The ACG 2024 guideline recommends eradication for all patients with confirmed H. pylori infection. Lack of symptoms alone is not an indication for observation.

Active infection not established: An isolated positive H. pylori IgG, particularly after previous therapy, does not establish current infection. Urea breath testing, stool antigen testing, or appropriate biopsy-based testing can establish whether active infection is present.

Temporary deferral: Eradication may be postponed when an acute competing illness, pregnancy without an urgent indication, significant drug interactions, or inability to complete treatment makes immediate therapy inappropriate. Reassess after the limiting condition resolves.

Repeated treatment failure: The AGA 2021 Clinical Practice Update (Best Practice Advice 9) recommends individualized discussion of additional eradication attempts, balancing benefits against adverse effects and repeated antibiotic exposure. Frailty, limited life expectancy, major intolerance, and absence of peptic ulcer disease or identifiable gastric cancer risk factors may favor withholding another empiric regimen. This is an expert-practice recommendation for refractory infection, not a routine alternative to initial treatment. Susceptibility testing may identify another usable regimen before further eradication attempts are abandoned.

References: ACG guideline 2024; AGA Clinical Practice Update 2021, Best Practice Advice 9 and 10.

Risks of untreated infection

Infection commonly persists for decades and causes chronic active gastritis. H. pylori is a class I gastric carcinogen and a cause of gastric MALT lymphoma. Eradication reduces, but does not abolish, subsequent gastric cancer risk, especially once premalignant mucosal changes are present. Individual risk depends on mucosal findings, family history, and regional background.

Resident core

2. Diagnostic testing and limitations

Urea breath testing (UBT) and validated monoclonal stool antigen testing are the usual noninvasive options when endoscopy is unnecessary. Gastric biopsies are appropriate when endoscopy is clinically indicated. Stool PCR may also provide clarithromycin resistance prediction.

Medication effects on UBT and stool antigen testing. Recent acid suppression, antibiotics, and bismuth can reduce detection on initial testing and after treatment. For example, Mayo’s breath-test pathway uses two weeks off PPIs and four weeks off antibiotics and bismuth. Vonoprazan can also suppress detection. H2 blockers or antacids may be an option during the washout, depending on the assay. A negative result during suppressive exposure may need repeating with the appropriate preparation. Stool PCR preparation is assay-specific: Mayo HPFRP specifies four weeks off antibiotics and two weeks off PPIs/bismuth. Collection and medication instructions differ among molecular assays; the performing laboratory's instructions govern the specific test.
TestBest useFalse negatives and other pitfalls
13C UBTInitial diagnosis and test of curePPIs, PCABs, antibiotics, bismuth, low organism burden; altered gastric anatomy can affect interpretation.
Monoclonal stool antigenInitial diagnosis and test of cureRecent suppressive medications; performance depends on assay. Laboratory monoclonal assays have stronger validation than many rapid tests.
Gastric histology and rapid urease testWhen EGD is indicated; simultaneously evaluate mucosal diseasePatchy colonization, gastric bleeding, atrophy, intestinal metaplasia, recent medications, inadequate antrum/body sampling. Negative biopsy does not always exclude infection.
Serologic IgGGenerally avoid in routine U.S. practiceAntibodies persist after eradication. A positive result is not evidence of current infection and cannot serve as test of cure.
Stool or biopsy PCRActive infection with selected resistance mutation detectionTarget-specific coverage, mutant heterogeneity, and recently treated infection; a negative resistance marker is not a complete susceptibility panel.
Biopsy culturePhenotypic susceptibility in refractory casesFastidious organism, transport delay, reduced recovery after therapy; negative culture cannot rule out infection.

A positive appropriately performed test for active infection generally establishes the diagnosis. The major medication-related problem is false negatives, not false positives. Rare UBT false positives can reflect urease-producing non-H. pylori flora or assay-specific circumstances. Discordant tests require review of timing, medications, assay, and sampling before labeling reinfection.

For treatment-naive adults with unknown susceptibility, the 2024 ACG guideline recommends optimized bismuth quadruple therapy for 14 days. Omeprazole-amoxicillin-rifabutin (Talicia) and vonoprazan-amoxicillin (Voquezna Dual Pak) are alternatives for patients without penicillin allergy.

RegimenComponents and product distinctionsEvidence and limitations
Optimized bismuth quadruple therapy
14 days
PPI + bismuth + tetracycline + metronidazole, usually prescribed as separate products. Pylera is a packaged bismuth regimen; Helidac is a historical co-pack.Preferred empiric option when susceptibility is unknown. Doxycycline is not a guideline-equivalent routine substitute for tetracycline. Adequate component exposure and completion are part of the definition of optimized therapy.
Rifabutin triple therapy
14 days
Omeprazole + amoxicillin + rifabutin. Talicia contains all three in a fixed-combination capsule.An empiric first-line alternative without penicillin allergy. Rifabutin triple therapy is also a salvage option after optimized BQT; first-line packaged-product evidence and salvage evidence are distinct.
Vonoprazan-amoxicillin dual therapy
14 days
Vonoprazan + amoxicillin, co-packaged as Voquezna Dual Pak.An empiric first-line alternative without penicillin allergy. First-line efficacy does not establish an equivalent role after repeated eradication failure.

Abbreviations. BQT, bismuth quadruple therapy; PCAB, potassium-competitive acid blocker; PPI, proton pump inhibitor. ACG 2024 regimen summary.

Talicia and separate-drug rifabutin regimens are not automatically interchangeable. Conventional rifabutin capsules differ in strength from the rifabutin component of Talicia; the formulation and complete regimen affect antibiotic exposure. The ACG regimen table and Talicia label describe the packaged regimen and separate-drug alternatives; these are not dose-equivalent by capsule substitution.

Pylera combines bismuth subcitrate potassium, metronidazole, and tetracycline; omeprazole is supplied separately. Its U.S. label describes a 10-day course with omeprazole for patients with H. pylori and active duodenal ulcer disease or a history of it within the past five years. This labeled population is narrower than ACG’s eradication guidance. Component exposure also differs from separate-drug 14-day BQT. The ACG guideline accepts the 10-day Pylera regimen; that duration alone does not establish inadequate treatment.

Clarithromycin- and levofloxacin-containing salvage regimens depend on demonstrated susceptibility. The fellow section distinguishes their role from empiric treatment.

Adherence, allergy, pregnancy, renal and hepatic function, interactions, and the feasibility of a complex regimen affect treatment selection. Fixed-combination capsules cannot be adjusted component by component.

Prior macrolide or fluoroquinolone exposure increases concern for resistance even when prescribed for an unrelated indication. Lack of exposure does not establish susceptibility.
Resident core

4. Test of cure

Confirmation of eradication relies on UBT or laboratory-based stool antigen testing at least four weeks after antibiotic treatment, with two weeks off PPIs or vonoprazan and four weeks off antibiotics or bismuth. Biopsy-based testing can be used if EGD is otherwise necessary. Symptom improvement is not proof of eradication, and serology cannot confirm cure.

Positive test of cure after an adequately timed, properly performed test generally indicates persistent infection. Without a documented previous negative test of cure, recurrence of a positive test does not establish reinfection rather than persistent infection or recrudescence.

With a negative appropriately timed test, persistent symptoms warrant evaluation for alternative causes rather than empiric repeat eradication therapy.

Fellow section

5. Management after treatment failure

Assessment of treatment failure includes the actual prior antibiotics, dose and frequency received, days completed, missed doses, acid suppression, prior macrolide or fluoroquinolone exposure, and timing of diagnostic tests. A regimen name alone does not distinguish resistance from inadequate exposure.

Prior treatmentSalvage considerations
Clarithromycin triple therapy or incomplete/nonoptimized treatment; no prior optimized BQTOptimized PPI + bismuth + tetracycline + metronidazole for 14 days is the guideline-supported empiric option. Pylera's 10-day course is also accepted by the guideline and is not inadequate solely because of its duration.
Completed optimized BQT without eradicationRifabutin triple therapy for 14 days is an empiric option without penicillin allergy; susceptibility-directed therapy may identify alternatives. Evidence for Talicia specifically after prior treatment is more limited than its first-line evidence. Formulation differences remain relevant.
Two failed therapies with confirmed adherence, or no usable empiric optionMolecular resistance testing and/or dedicated biopsy culture can help select subsequent therapy. This also requires review of adherence, acid suppression, prior exposure, and whether further attempts offer sufficient benefit.
Known clarithromycin or levofloxacin susceptibilityClarithromycin combinations pair amoxicillin with a PPI or vonoprazan. Voquezna Triple Pak has no ACG recommendation for or against salvage use because of insufficient evidence; Table 6 also excludes prior macrolide exposure. Levofloxacin triple therapy combines a PPI + amoxicillin + levofloxacin for 14 days and is reserved for susceptible infection when optimized BQT or rifabutin therapy has already been used or is unavailable. Product approval alone does not establish the appropriate salvage sequence.

Abbreviations. BQT, bismuth quadruple therapy; EGD, esophagogastroduodenoscopy; PCAB, potassium-competitive acid blocker; PPI, proton pump inhibitor. ACG 2024; AGA refractory-infection advice.

After failed PPI-clarithromycin triple therapy, ACG limits repeat use to confirmed susceptibility to both clarithromycin and amoxicillin, with an opportunity to improve acid suppression after adherence review. This is expert consensus. Clarithromycin or levofloxacin susceptibility is necessary for a susceptibility-directed salvage regimen but does not guarantee eradication.

ACG describes a narrow treatment-naive exception for vonoprazan-clarithromycin triple therapy when susceptibility testing and other first-line options are unavailable, without prior macrolide exposure or penicillin allergy. This does not establish an empiric salvage role.

Clarithromycin resistance commonly follows 23S rRNA mutations and predicts poor results with clarithromycin combinations. Fluoroquinolone resistance commonly follows gyrA alterations. Metronidazole resistance is more complex; adequate metronidazole exposure with bismuth and acid suppression can sometimes overcome in vitro resistance. Amoxicillin, tetracycline, and rifabutin resistance are generally less prevalent, but susceptibility patterns vary geographically.

Fellow section

6. Molecular resistance testing and culture

Send-out assay examples were checked against the public catalogs on November 9, 2026. They illustrate different testing approaches. Local availability, ordering, and coverage depend on the submitting laboratory and payer. Mayo Clinic Laboratories HPFRP (stool) and HPRP (tissue) detect H. pylori and selected 23S rRNA clarithromycin resistance mutations. They do not constitute broad antibiotic susceptibility panels. American Molecular Laboratories (AML) lists a broader next-generation sequencing panel, PyloriDx/PyloriAR, for H. pylori detection and prediction of resistance to six agents. Molecular assays predict resistance from detected variants; they do not measure minimum inhibitory concentrations or prove susceptibility to agents without a detected resistance marker.

AML resistance markerAntibiotic or class
23S rRNAClarithromycin
gyrAFluoroquinolones, including levofloxacin
rdxAMetronidazole
pbp1Amoxicillin
16S rRNATetracycline
rpoBRifabutin

Abbreviations. AML, American Molecular Laboratories; NGS, next-generation sequencing; PCR, polymerase chain reaction.

Prediction is better characterized for clarithromycin and levofloxacin than for metronidazole, for which genotype-to-phenotype concordance remains problematic. Evidence for prediction of amoxicillin, tetracycline, and rifabutin resistance is limited by the low prevalence of resistant isolates. Prior treatment and the specific assay validation affect interpretation of each reported marker.

AML PyloriDx/PyloriAR send-out and requisition

AML H. pylori detection and six-drug resistance NGS panel accepts formalin-fixed paraffin-embedded (FFPE) tissue, fresh gastric tissue, or stool according to the laboratory's published service description. This may permit testing archived gastric pathology tissue without repeat endoscopy when an adequate block is available. Stool offers a noninvasive option; washout and specimen handling requirements remain assay-specific.

The AML forms page lists a test requisition. Archived tissue may permit resistance testing without repeat EGD, but adequacy of the block, pathology release, handling, access, and coverage require coordination with the receiving laboratory. This is a send-out option rather than a verified local order pathway.

Culture requires a suitable unfixed gastric specimen. When biopsy is used, culture tissue is collected separately from routine formalin-fixed pathology under the receiving laboratory's transport conditions. A routine formalin-fixed specimen for histopathology will not produce a bacterial culture. A negative culture following therapy is common enough that an unsuccessful recovery must not be mistaken for susceptibility.

Mayo clarithromycin PCR and culture send-outs

HELIS, H. pylori culture with antimicrobial susceptibilities, requires a suitable unfixed specimen and rapid transport. The catalog defines acceptable containers, temperature, and timing. Dedicated culture tissue is separate from routine formalin-fixed pathology. Rifampin susceptibility, when reported, is not a directly measured rifabutin MIC.

HPFRP, stool PCR, and HPRP, tissue PCR, predict clarithromycin resistance. HPRP accepts specified fresh or FFPE tissue; its specimen requirements differ from culture. The Mayo diagnostic algorithm links these pathways. Preprocedure coordination helps prevent loss of the opportunity for culture when all tissue is placed in formalin.

Fellow section

7. Treatment considerations in special populations

SituationClinical considerations
True immediate penicillin allergyOptimized BQT is usually the principal empiric option. Formal allergy evaluation can restore effective amoxicillin-containing salvage options; a reported childhood reaction does not necessarily represent persistent allergy.
CKD, especially severe impairmentTalicia labeling advises avoidance when GFR is below 30 mL/min. Voquezna Dual Pak and Triple Pak are not recommended when eGFR is below 30 mL/min. Pylera is contraindicated in severe renal impairment. Separate-drug regimens require component-specific dosing and contraindication review; fixed-combination capsules do not permit independent component adjustments; co-packs have their own labeling restrictions.
Pregnancy or breastfeedingNonurgent eradication may be deferred during pregnancy when clinically appropriate. Pylera is contraindicated during pregnancy. An urgent indication requires GI/obstetric review and an individualized pregnancy-compatible regimen. Breastfeeding requires separate product-specific advice; do not apply pregnancy restrictions indiscriminately to lactation.
Prior severe antibiotic-associated colitis or C. difficileRecurrence risk and stewardship tradeoffs are relevant, but prior C. difficile alone does not remove the potential benefit of eradication.
Hepatic impairment, cytopenias, interacting medicationsTalicia labeling advises avoidance in hepatic impairment, including Child-Pugh A, B, or C. Voquezna Dual Pak and Triple Pak are not recommended in Child-Pugh B or C. Pylera is not recommended in severe hepatic impairment. Metronidazole exposure, rifabutin interactions/myelotoxicity, and clarithromycin QT/CYP3A interactions warrant separate consideration. Baseline CBC may be appropriate for rifabutin in high-risk patients.
Immunosuppressive therapyRifamycins and clarithromycin have consequential interactions with transplant/immunosuppressive drugs; immunosuppression itself does not establish a different eradication regimen.

Product-specific restrictions: Talicia US prescribing information, sections 8.6–8.7; Voquezna packs US prescribing information, sections 2.3–2.4 and 8.6–8.7; Pylera US prescribing information, sections 4.4–4.5 and 8.7.

Clinical reference

Approximate test accuracy and interpretation

Performance depends on the assay, reference standard, sampling, population, and suppressive medications. The source-specific estimates below are not head-to-head comparisons or substitutes for the performing laboratory's validation. The stool-antigen row identifies a historical pooled estimate; the UBT and urease rows summarize ranges or estimates reported in the linked diagnostic review.

ModalitySensitivitySpecificityImportant caveat
13C UBT94–99%93–100%Range summarized by Schuetz et al., 2024; assay and comparator dependent.
Monoclonal stool antigen, before treatment94% (95% CI 93–95%)97% (95% CI 96–98%)Gisbert et al., 2006 meta-analysis: 22 studies, 2,499 patients. Historical pooled performance does not validate every current rapid assay.
HistopathologySampling and method dependentStain and interpretation dependentNo single transferable estimate. Sample antrum and corpus; patchy distribution, low organism burden and prior suppression can reduce detection.
Rapid urease biopsyAbout 90%At least 95%Estimate summarized by Schuetz et al., 2024; sampling and organism burden matter, and performance after treatment is less reliable.
SerologyHighly assay-dependentHighly assay-dependentEven an analytically accurate IgG assay cannot distinguish past from active infection.
CultureVariable, often below molecular methodsVery high for organism identificationRecovery heavily affected by transport and prior treatment.

No universal sensitivity/specificity should be assigned to stool PCR: use the specific assay and target validation. Sources: Schuetz et al., 2024 diagnostic review; Gisbert et al., 2006 monoclonal stool-antigen meta-analysis.

Commercial combination products
ProductComponents and labeled courseDistinction
PyleraBismuth subcitrate potassium + metronidazole + tetracycline in a fixed-combination capsule, with separate omeprazole; 10 days.The U.S. label covers patients with active duodenal ulcer disease or a history within five years. ACG accepts the 10-day packaged regimen. Its exposure differs from separate-drug 14-day BQT.
Helidac, historical co-packBismuth subsalicylate + metronidazole + tetracycline; historical labeling described 14 days with an H2-receptor antagonist.The brand is no longer marketed. ACG does not recommend H2-receptor antagonists as the acid suppressant in eradication regimens.
TaliciaOmeprazole + amoxicillin + rifabutin in a fixed-combination capsule; 14 days.Conventional rifabutin capsules do not reproduce the packaged formulation by capsule-for-capsule substitution.
Voquezna Dual PakVonoprazan + amoxicillin co-pack; 14 days.Empiric first-line alternative without penicillin allergy; salvage evidence remains limited.
Voquezna Triple PakVonoprazan + amoxicillin + clarithromycin co-pack; 14 days.Clarithromycin susceptibility and prior-treatment history determine suitability. ACG makes no salvage recommendation because of insufficient evidence and excludes prior macrolide exposure in that setting.

Product names identify different components, strengths, and schedules. Multiple daily administrations and pill burden can affect completion. Current labels provide the prescribing details.

Pylera label; Helidac market status; Talicia label; Voquezna pack labels.

Adverse effects and medication counseling

Bismuth can darken stool or tongue and may be unsuitable with salicylate hypersensitivity (subsalicylate formulation). Tetracycline interacts with polyvalent cations and is avoided in pregnancy; coadministration with calcium, iron, and antacids can reduce absorption. Metronidazole causes nausea, metallic taste and neuropathy with prolonged exposure; warfarin interaction can increase anticoagulant effect. Rifabutin can cause cytopenias, uveitis, discoloration of secretions, CYP interactions and resistance concerns when used for other mycobacterial indications. Amoxicillin requires allergy and renal assessment. Clarithromycin has QT, CYP3A, and drug-interaction concerns. Levofloxacin carries tendon, neuropathy, CNS, dysglycemia, aortic aneurysm risk and other FDA warnings; renal adjustment is required. Vonoprazan has drug interactions and renal/hepatic labeling considerations.

The examples below illustrate clinically important product interactions; current labeling contains the complete restrictions. Talicia is contraindicated with rilpivirine, delavirdine, or voriconazole and can reduce hormonal contraceptive efficacy; its label advises additional nonhormonal contraception. Both Voquezna packs are contraindicated with rilpivirine. Triple Pak adds clarithromycin restrictions, including contraindicated simvastatin/lovastatin coadministration and colchicine in patients with renal or hepatic impairment, and should be avoided in patients with known QT prolongation. These examples are not a complete interaction screen. Talicia labeling; Voquezna pack labeling.

Clinical cases

Case 1. A patient took 14 days of BQT. A stool antigen assay is negative 8 days after finishing treatment while the patient is still taking pantoprazole. Discussion. This result cannot reliably establish cure because the test was too early and acid suppression continued. A properly timed test after the relevant washout is needed to assess eradication.

Case 2. A patient failed PPI + amoxicillin + clarithromycin, then reports taking bismuth + tetracycline + metronidazole + PPI (“quadruple therapy”) but only received 7 days and missed many afternoon doses. Discussion. The documented exposure does not establish failure of optimized BQT. Regimen names alone cannot diagnose multidrug resistance.

Case 3. Two fully documented regimens failed and stool PCR predicts clarithromycin resistance. Discussion. The resistance result argues against a clarithromycin-containing regimen. Adherence, previous exposure, and broader susceptibility information may guide the next treatment; dedicated biopsy culture is one option.

References

Chey WD et al. ACG Clinical Guideline: Treatment of Helicobacter pylori Infection. Am J Gastroenterol. 2024;119:1730–1753. Full guideline; ACG summary.

Shah SC, Iyer PG, Moss SF. AGA Clinical Practice Update on the Management of Refractory Helicobacter pylori Infection: Expert Review. Gastroenterology. 2021;160:1831–1841. Best Practice Advice 9–10.

Malfertheiner P et al. Management of Helicobacter pylori infection: Maastricht VI/Florence consensus report. Gut. 2022. Consensus.

American Molecular Laboratories. PyloriDx/PyloriAR six-drug NGS panel; specimen requisition and forms. See also Testing for H. pylori in an era of antimicrobial resistance, J Clin Microbiol, 2024.

Mayo Clinic Laboratories. H. pylori diagnostic algorithm, December 2025; HELIS culture and susceptibility instructions.

Product labeling provides current prescribing details and restrictions. Laboratory test menus, send-out access, and susceptibility patterns vary.