Helicobacter pylori is a bacterium that colonises the lining of the stomach, usually in childhood, and then stays for life unless it is treated. Most people who carry it have no symptoms. A minority develop ulcers, and a smaller minority develop gastric cancer.
It is the only bacterium classified by the International Agency for Research on Cancer as a Group 1 carcinogen, meaning there is sufficient evidence that it causes cancer in humans. That classification was made in 1994, and the evidence has only strengthened since.
How common it is
Global prevalence has fallen substantially, from around 58% in the 1980s to around 43% in the period from 2011 to 2022. In Western Europe it is around 34%, and Southern and Eastern Europe sit significantly higher.
The decline reflects better sanitation and living conditions across generations. It also means that prevalence in older adults, who grew up under different conditions, remains higher than the population average.
The link with gastric cancer
The connection is stronger than most patients expect.
The International Agency for Research on Cancer estimates that 89% of non-cardia gastric cancers are attributable to H. pylori. Modelling published in 2025 estimated that of the gastric cancers expected across the lifetime of people born between 2008 and 2017, 76% are attributable to this infection, and up to three quarters could be prevented by screening and treating for it.
Crucially, eradication changes the outcome. A meta-analysis of randomised trials found that eradicating H. pylori in infected but otherwise healthy people reduced gastric cancer incidence by around 36% and gastric cancer mortality by around 22%.
This is unusual in oncology. A single course of antibiotics, in the right person, measurably reduces the risk of a cancer that is otherwise often diagnosed late.
Who should be tested
Testing is indicated, rather than optional, in the following situations:
- Current or previous peptic ulcer disease
- Gastric MALT lymphoma, where eradication alone produces long-term remission in 70% to 80% of cases
- Dyspepsia in a patient without alarm features, below the age threshold for endoscopy
- Functional dyspepsia
- Gastric premalignant conditions such as atrophic gastritis or intestinal metaplasia, or previous gastric cancer
- A first-degree relative with gastric cancer
- Before starting long-term NSAID treatment, which reduces the risk of ulcer
- Unexplained iron deficiency anaemia
- Immune thrombocytopenic purpura, where platelet counts rise after eradication
- Long-term proton pump inhibitor use, since untreated infection under long-term PPI shifts gastritis towards the body of the stomach and increases atrophy
For low dose aspirin, testing is recommended only in patients already at high risk of ulcer, because the number needed to treat in the general aspirin-using population is very high.
The test and treat approach
For a younger patient with dyspepsia and no alarm features, the approach is to test non-invasively and treat if positive, rather than proceed straight to endoscopy. European consensus places the age threshold at around 50, with a range of 45 to 55 depending on local cancer incidence. Below that threshold the prevalence of gastric cancer among patients with dyspepsia is around 0.4%.
Above that age, or in the presence of alarm features such as weight loss, vomiting, difficulty swallowing, bleeding or anaemia, endoscopy comes first.
Which test to use
The urea breath test is the most accurate non-invasive option. You swallow a solution containing urea labelled with carbon-13. If H. pylori is present, the enzyme it produces splits the urea and releases labelled carbon dioxide, which is then measured in your breath. No bacteria, no labelled CO2.
In pooled analysis of 39 studies, the carbon-13 breath test showed a sensitivity of 96.6% and a specificity of 96.9%. The carbon-14 version performs similarly for sensitivity but is less specific and involves a small dose of radiation, which is why the carbon-13 version is preferred.
Stool antigen testing is the other reliable non-invasive option, with comparable accuracy in validated laboratory formats.
Blood antibody testing should not be used to diagnose active infection, and must never be used to confirm eradication. Antibodies persist for years after the bacterium has gone, so a positive result cannot distinguish current infection from one treated a decade ago.
Biopsy-based testing during gastroscopy includes the rapid urease test, histology and culture. Where a gastroscopy is being performed anyway, biopsies allow both diagnosis and assessment of the stomach lining. European guidance now also supports molecular testing for clarithromycin resistance mutations.
Preparing for the breath test
These rules are not optional. Ignoring them produces a false negative, and a false negative means an infection stays untreated.
- Stop proton pump inhibitors at least two weeks before the test. These are the medicines most people are taking for the symptoms that prompted the test in the first place, which is exactly why this is missed so often.
- Stop antibiotics and bismuth preparations at least four weeks before.
- H2 receptor antagonists do not interfere with the test, so they are an option for symptom relief during the washout period. Discuss this with your doctor rather than simply stopping everything.
- Fasting is required before the test. Follow the specific instruction given by the clinic performing it, as protocols vary.
Treatment in Greece, and why antibiotic resistance matters
Treatment is not a single standard course everywhere in the world, because resistance patterns differ by region.
Across Europe, resistance in the most recent period has been reported at approximately 24% for clarithromycin, 33% for metronidazole and 21% for levofloxacin, with markedly higher rates in Southern European countries. Greek studies describe Greece as a high clarithromycin resistance area, above 20%.
That figure matters because the guideline threshold is 15%. Above it, the standard clarithromycin-based triple therapy that many patients remember from years ago is no longer reliable first-line treatment.
Current European guidance for high resistance areas recommends bismuth quadruple therapy as first line, or non-bismuth concomitant quadruple therapy where bismuth is not available. In a Greek trial, quadruple regimens achieved eradication in around 88% to 92% of patients.
Vonoprazan-based regimens, used in the United States and parts of Asia, are not available in the European Union at the time of writing.
The practical consequence for patients: do not assume the treatment a relative received ten years ago is the right treatment now, and complete the full course exactly as prescribed. Incomplete treatment is one of the main drivers of resistance.
Confirming the infection is gone
This step is skipped far too often, and it should not be.
- Retest at least four weeks after finishing treatment, allowing a drug-free interval so that suppressed bacteria are not mistaken for eradicated ones
- Use a urea breath test or stool antigen test, never serology
- Stop PPIs for at least two weeks before the test of cure, as for the initial test
Reinfection after successful eradication is uncommon, at roughly 1% per year.
When to see a doctor
Arrange an assessment if you have:
- Persistent upper abdominal pain or burning, particularly if it wakes you at night or improves with food
- Dyspepsia that keeps returning after you stop antacids
- Nausea, early satiety or unexplained loss of appetite
- A first-degree relative with gastric cancer
- Iron deficiency anaemia with no obvious cause
- A history of peptic ulcer that was never confirmed as eradicated
And seek assessment promptly, with endoscopy rather than a breath test, if you have difficulty swallowing, vomiting, weight loss, or evidence of bleeding such as black stools or vomiting blood.