How Accurate Is the H. Pylori Breath Test?

The urea breath test for Helicobacter pylori is one of the more accurate non-invasive diagnostic tools in gastroenterology, with pooled sensitivity around 96% and specificity around 93% across large meta-analyses of patients with digestive symptoms.¹ That said, those headline numbers come with important caveats. Certain medications, medical conditions, and even the timing of the test can push accuracy well below that range, and borderline results are more common than most people expect.

What the Large Studies Actually Show

Two major meta-analyses give a good picture of how the breath test performs across diverse clinical settings. One, pooling data from studies that used both carbon-13 and carbon-14 labeled urea, reported sensitivity estimates between about 88% and 93% and specificity between roughly 85% and 90%, depending on the statistical model used. That same review noted wide variation among individual studies, with sensitivity in some ranging as low as 64% and specificity as low as 61%.² A second meta-analysis, focused on patients with dyspepsia, found tighter numbers: pooled sensitivity of 96% and specificity of 93%.¹

The gap between those two sets of numbers matters. The breath test performs best under clean conditions: the patient has fasted, hasn’t recently taken certain medications, and has a mostly intact stomach. When studies include messier real-world populations, accuracy dips. So while the test is genuinely good, how good depends on whether you and your doctor got the preparation right.

How the Test Works

H. pylori produces large amounts of an enzyme called urease, which breaks down urea. The breath test exploits this by having you swallow a solution containing urea tagged with a traceable carbon isotope (either carbon-13 or carbon-14). If H. pylori is living in your stomach lining, its urease splits the labeled urea, releasing labeled carbon dioxide. That CO₂ is absorbed into the bloodstream, carried to the lungs, and exhaled. A breath sample collected before and after drinking the solution is analyzed for the isotope; if the labeled CO₂ rises above a threshold, the test is positive.³

Carbon-13 is a stable, non-radioactive isotope and is the version most widely used today, especially for children and pregnant women. Carbon-14 is mildly radioactive, though the dose is tiny. Both versions perform comparably in clinical studies. The machines used to analyze breath samples have also been validated against each other: infrared spectrometers and traditional mass spectrometers produce nearly identical results, with sensitivity ranging from 97% to 100% in head-to-head comparisons.⁴

The Biggest Threat to Accuracy: Proton Pump Inhibitors

If you take a proton pump inhibitor (PPI) like omeprazole, lansoprazole, or esomeprazole, your breath test result could be wrong. PPIs don’t just reduce acid; they also suppress H. pylori enough to make the bacterium temporarily undetectable without actually killing it. In one study, about a third of patients whose infection had not been eradicated showed a falsely negative breath test while on lansoprazole.⁵ Another study found even worse numbers: false negatives in 45% of infected patients after four weeks of high-dose omeprazole, and in 28% after six months of standard-dose treatment. The same study also found false positives in 15% of uninfected patients on high-dose omeprazole, and it took two to four weeks after stopping treatment for the test’s accuracy to recover.⁶

This is why gastroenterology guidelines uniformly recommend stopping PPIs at least two weeks before a breath test. Antibiotics need an even longer washout: four weeks minimum, because they can suppress bacterial populations for extended periods without fully clearing the infection. The Maastricht VI/Florence consensus and the 2024 American College of Gastroenterology guidelines both reinforce these timing rules.⁷ If your doctor orders a breath test and nobody asks about your current medications, that is a red flag worth raising yourself.

False Positives and What Causes Them

False positives on the breath test are less common than false negatives, but they do happen. The main culprit is other bacteria in the mouth or stomach that also produce urease. In one study of 102 patients, four had false-positive results, and in each case, urease-producing bacteria other than H. pylori were isolated from the oral cavity or stomach. All four of those patients had atrophic gastritis, a condition where the stomach lining thins and acid production drops, which may allow non-H. pylori bacteria to colonize the stomach more easily.⁸

This is a relatively uncommon scenario for most people, but it becomes more relevant for older adults or anyone with known atrophic gastritis. If you receive a positive breath test and your doctor has reason to suspect atrophic changes, a confirmatory test through endoscopy or stool antigen may be worthwhile.

The Gray Zone: When Results Fall Near the Cutoff

The breath test produces a numerical value, not just a yes or no. A reading above a set threshold (commonly 4‰ for the carbon-13 version) is called positive. A reading below a lower threshold (often 2‰) is called negative. But readings that land in between sit in a gray zone where the test is significantly less reliable.

A study that specifically examined this gray zone found that among patients whose results fell between 2‰ and 4‰, the false-negative rate was substantial: about 29% in the 2‰–3‰ range and nearly 49% in the 3‰–4‰ range. On the other side, among those in the 4‰–5‰ range, roughly 20% were false positives. The reliability improved once results climbed above about 5‰, where false positives dropped below 7%.⁹ The researchers also identified intestinal metaplasia in the stomach antrum as an independent risk factor for false-negative results in this borderline range.

The practical takeaway: if your breath test value lands close to the cutoff, the result should be interpreted cautiously. A follow-up study recommended either repeating the breath test or confirming with another diagnostic method when values hover near the cutoff.¹⁰ If your result report includes the actual numerical value and it sits near the decision line, ask your doctor whether a second test is warranted.

How Accuracy Changes After Stomach Surgery

For people who have had part of their stomach removed (partial gastrectomy), the breath test becomes less reliable. The test depends on the labeled urea spending enough time in contact with the stomach lining for H. pylori’s urease to act on it. After gastrectomy, the stomach empties faster, the available mucosal surface is smaller, and the anatomy is altered. One study concluded plainly that the breath test is less effective in post-gastrectomy patients than in those with intact stomachs, and less effective than endoscopy for this group.¹¹

Meta-analyses bear this out. One pooled analysis of nine studies found sensitivity of only about 77% and specificity of 89% in post-gastrectomy patients.¹² A more recent meta-analysis reported a pooled sensitivity of 83% and specificity of 79%, with individual study sensitivities ranging as low as 40%.¹³ If you have had gastric surgery, endoscopy with biopsy is generally preferred for H. pylori diagnosis.

Accuracy in Children

The breath test works well in children, but the standard adult cutoff values may not be ideal for younger kids. One study found that children aged six and under had a false-positive rate of about 8% using the standard cutoff, compared with less than 1% in older children. Adjusting the threshold upward for the younger group eliminated most of those false positives without sacrificing sensitivity.¹⁴ A separate study in a mixed pediatric population found that using a slightly higher cutoff at the 30-minute mark yielded sensitivity of 95% and specificity of 97%.¹⁵

The bottom line for parents: the test is reliable in children, but the lab and clinician should ideally be using age-appropriate cutoff values. If your child gets a positive result and they are under six, it is worth confirming that the result wasn’t a false positive driven by the standard adult threshold.

When Ulcers Are Actively Bleeding

Testing for H. pylori during an acute bleeding episode is tricky. Patients with peptic ulcer bleeding are often fasting, may be on intravenous PPIs, and the bleeding itself can alter the stomach environment. One systematic review noted that the breath test has very low feasibility in this acute setting because of the fasting requirements around endoscopy.¹⁶

That said, when the test can be performed shortly after feeding resumes, it still performs reasonably. One study found that most infected patients with ulcer bleeding had a positive breath test done right after they resumed eating, even after high-dose PPI treatment. But the researchers cautioned that if the first test comes back negative, a second test should be done after PPIs have been stopped, to rule out a drug-induced false negative.¹⁷ A prospective trial that compared methods found the first-day breath test had 100% sensitivity in bleeding ulcer patients, outperforming histology and rapid urease testing done from biopsies.¹⁸

How It Compares to Stool Antigen and Blood Tests

The breath test is not the only non-invasive option. The stool antigen test and blood serology (antibody test) are both widely available, and how they stack up depends on what you are trying to learn.

In a head-to-head comparison in children, the breath test and the stool antigen test performed nearly identically: the breath test had sensitivity of 95%, specificity of 98%, and accuracy of 96%, while the stool antigen test came in at 98% sensitivity, 98% specificity, and 98% accuracy. The differences were not statistically significant.¹⁹ In adult populations, the stool antigen test shows similarly competitive numbers, and guidelines generally endorse either as a first-line non-invasive test.

Blood serology is a different story. It detects antibodies to H. pylori, which means it can tell you whether you have ever been infected but cannot distinguish between a current active infection and one that was successfully treated months or years ago. One study found serology had high sensitivity (about 97%) and a very high negative predictive value (about 98%), meaning a negative blood test is quite reliable at ruling infection out. But its specificity was only about 79%, meaning roughly one in five positives would be from someone with past rather than current infection.²⁰ This is why a positive blood test often needs confirmation with a breath test or stool antigen test. If you need to know whether the infection is active right now, the breath test or stool antigen test is the better choice.

Preparation Tips That Affect Your Result

Most clinics will tell you to fast for a certain number of hours before the breath test. One large eradication-confirmation study reported patients fasting for an average of about five hours, with a range from three to nine hours.²¹ The fast prevents food from diluting or interfering with the urea solution.

Some testing protocols ask you to drink a citric acid solution before or with the labeled urea. Citric acid slows gastric emptying, which keeps the urea in contact with the stomach lining longer and improves the signal.²² A study comparing different test meals found that using a citric acid drink produced a stronger and earlier signal than semi-liquid meals: peak values were reached at around 30 minutes with citric acid versus 45 to 60 minutes with food-based test meals, and sensitivity hit 96–100% within 15 minutes with citric acid.²³ Not all labs use citric acid, though. If your protocol doesn’t include it, the test still works, just with a slightly longer collection window.

Using the Breath Test to Confirm Treatment Worked

One of the most common reasons for ordering a breath test isn’t initial diagnosis but confirming that H. pylori has been successfully eradicated after treatment. This is where timing matters enormously. Testing too soon after finishing antibiotics and PPIs risks a false negative, because residual drug effects can suppress the bacterium just enough to hide it.

Guidelines recommend waiting at least four weeks after completing treatment before testing. One study that checked breath test accuracy just two weeks after therapy found sensitivity of 90%, specificity of 99%, and overall accuracy of 97%, which was comparable to results at four to six weeks.²⁴ Still, most current guidelines err on the side of caution and stick with four weeks as the minimum. In practice, many patients are tested around seven to eight weeks post-treatment.

In a recent eradication-confirmation study, about 76% of patients had negative breath tests after first-line therapy, while roughly 24% tested positive, indicating treatment failure.²¹ Those treatment-failure patients would then typically receive a second-line antibiotic regimen, followed by another breath test to confirm eradication. The test’s non-invasive nature makes it well suited for this repeated-testing role.

Cost and Accessibility in Primary Care

The breath test is more expensive per unit than a simple blood antibody test, but it avoids the downstream costs of unnecessary endoscopies and misdiagnoses. A UK-based economic analysis found that a “test-and-treat” approach using either the breath test or stool antigen test was far more cost-effective for managing dyspepsia than jumping straight to endoscopy.²⁵ A German health-technology assessment reached similar conclusions, finding that the breath-test-based test-and-treat strategy was cost-effective over serology-based approaches in the majority of models analyzed and consistently cheaper than endoscopy-first strategies.²⁶

An Ontario analysis offered a different angle: when comparing the breath test head-to-head against serology alone, the breath test was more expensive and not always more effective in population-level modeling, because the higher per-test cost adds up at scale. A two-step strategy of serology followed by confirmatory breath testing for positive results emerged as the most efficient option in that analysis.²⁷ So the cost picture depends on the clinical context: for initial screening in low-prevalence populations, a cheaper screening test followed by breath-test confirmation can make sense. For confirming eradication after treatment, where you specifically need to know about active infection, the breath test is the standard.

When the Breath Test Outperforms Endoscopy

Endoscopy with biopsy is often treated as the gold standard for H. pylori diagnosis, but the breath test can actually be more accurate in some settings. An endoscopic breath test study found that the breath test’s accuracy (98–100%) exceeded that of culture (88%) and the rapid urease biopsy test (92–96%), with the differences reaching statistical significance for some comparisons.²⁸ Biopsy-based tests can miss patchy infections because they sample only a few spots, while the breath test reflects the total bacterial load across the entire stomach. This whole-stomach sampling is one reason why the breath test’s sensitivity remains competitive even against invasive methods.