The breath test is the workhorse of SIBO diagnosis. If you have ever been tested for small intestinal bacterial overgrowth, you probably drank a solution of glucose or lactulose, then breathed into a collection device every 15 to 20 minutes for two to three hours. The idea is straightforward: bacteria in the small intestine ferment the sugar substrate and produce gases that are absorbed into the bloodstream, carried to the lungs, and exhaled. Measure the gases, and you can infer whether there is excessive fermentation happening where it should not be. The concept is elegant. The execution, unfortunately, is where things get complicated. Breath tests are non-invasive, widely available, and relatively inexpensive, but their accuracy has been questioned since they were first developed. Understanding what they actually measure, what they miss, and what might replace them is essential for anyone navigating a SIBO diagnosis.
The three gases and what they tell us
Human cells do not produce hydrogen, methane, or hydrogen sulfide. When these gases appear in exhaled breath, they come from microbial fermentation in the gut. This is the biological basis that makes breath testing possible, and it is well established (Levitt, 1969). But the three gases come from different organisms and have different clinical associations.
Hydrogen is produced by a wide range of bacteria during carbohydrate fermentation. An early rise in hydrogen during a breath test (within the first 90 minutes for glucose, or before the expected colonic arrival time for lactulose) suggests small intestinal fermentation. The 2017 North American Consensus defined a positive hydrogen result as an increase of 20 or more parts per million above baseline within 90 minutes of substrate ingestion (Rezaie et al., 2017). Clinically, hydrogen-dominant SIBO tends to be associated with diarrhea-predominant symptoms.
Methane is produced not by bacteria but by archaea, primarily Methanobrevibacter smithii, which consume hydrogen generated by other organisms and produce methane as a byproduct. Methane slows intestinal transit time directly by acting on smooth muscle and increasing non-propagating contractions (Pimentel et al., 2006). This is why methane-positive patients tend toward constipation. The consensus threshold is 10 or more parts per million at any point during the test. Because methanogens operate throughout the intestinal tract, methane elevation now falls under the term IMO (intestinal methanogen overgrowth) rather than SIBO.
Hydrogen sulfide is the most recently recognized gas in breath testing. Produced by sulfate-reducing bacteria such as Desulfovibrio and Bilophila species, hydrogen sulfide has been associated with diarrhea-predominant symptoms, possibly through its effects on intestinal motility, mucosal inflammation, and visceral sensitivity (Singer-Englar et al., 2022). Until recently, most breath testing devices could not measure hydrogen sulfide, meaning an entire category of patients was functionally invisible to standard testing.
The 15 to 30 percent problem: patients who produce mainly methane or hydrogen sulfide
For decades, the standard breath test measured only hydrogen. If your gut microbiome happened to be dominated by methanogenic archaea or sulfate-reducing bacteria rather than hydrogen-producing organisms, your breath test would come back flat. No hydrogen rise. Test negative. Go home. This was a significant clinical problem because an estimated 15 to 30% of the population falls into this category (Pimentel et al., 2020).
The addition of methane measurement to standard breath testing helped considerably. Two-gas testing became the norm at most major gastroenterology centers by the mid-2010s, and it captured the roughly 15 to 20% of patients with methane-dominant overgrowth who would otherwise have been missed. But hydrogen sulfide producers remained undetectable until the development of three-gas testing devices.
The TrioSmart device, developed by a team led by Pimentel and colleagues at Cedars-Sinai, became the first FDA-cleared breath test to measure hydrogen, methane, and hydrogen sulfide simultaneously. Early data suggested that adding hydrogen sulfide measurement identified an additional subset of patients with diarrhea-predominant symptoms whose two-gas tests had been negative (Singer-Englar et al., 2022). This is a meaningful advance, though the clinical protocols for treating hydrogen sulfide-dominant overgrowth are still being developed.
âšī¸If you have had a two-gas breath test that came back negative but you still have significant symptoms, it may be worth discussing three-gas testing with your gastroenterologist. You could be a primarily hydrogen sulfide producer who was missed by the older test.
Sensitivity and specificity: where the numbers actually land
The accuracy of any diagnostic test is measured by its sensitivity (how well it catches true positives) and specificity (how well it avoids false positives). For breath tests, these numbers are, to put it plainly, not great. A systematic review by Khoshini and colleagues found that the sensitivity of the glucose breath test ranged from 20 to 93% and the lactulose breath test from 31 to 68% when compared to jejunal aspirate culture (Khoshini et al., 2008). That is an enormous range, and it reflects differences in study design, patient populations, and how the reference standard was applied.
Glucose breath tests tend to have higher specificity (fewer false positives) because glucose is absorbed in the proximal small intestine, so any gas production before absorption is completed likely reflects small intestinal bacteria. The trade-off is that glucose does not reach the distal small bowel, so overgrowth confined to the ileum will be missed entirely. Lactulose, being non-absorbable, traverses the entire small intestine, which gives it potentially better sensitivity for distal overgrowth. But lactulose inevitably reaches the colon, where it is fermented by normal colonic flora, making it very difficult to distinguish small intestinal from colonic gas production based on timing alone.
The practical implication is that a positive breath test increases the probability of SIBO but does not confirm it, and a negative test decreases the probability but does not exclude it. Clinicians who understand these limitations use breath tests as one piece of a larger clinical picture rather than treating the result as definitive.
Jejunal aspirate: the gold standard and its problems
The reference standard for SIBO diagnosis is a quantitative culture of fluid aspirated from the jejunum (the middle section of the small intestine) during upper endoscopy. A bacterial count exceeding 10^3 colony-forming units per milliliter is considered diagnostic, though some experts and older literature used a higher threshold of 10^5 CFU/mL (Quigley et al., 2020).
On paper, this is the definitive test. In practice, it has several significant limitations. First, the aspirate samples only one location in the small bowel, and bacterial distribution may be patchy. Second, oropharyngeal contamination during the endoscopy procedure can introduce bacteria that were not originally in the small intestine, producing false positives. Third, not all bacteria grow well in standard culture conditions, meaning some species may be present but uncounted. Fourth, there is no universally standardized protocol for how the aspirate should be collected, transported, or cultured, so results can vary between laboratories. And fifth, the procedure is invasive, requires sedation, and is considerably more expensive than a breath test.
These limitations mean that while jejunal aspirate is the best reference standard we have, it is an imperfect one. When breath tests are validated against jejunal aspirate, both the sensitivity and specificity estimates are influenced by the aspirate's own error rate, which is difficult to quantify.
Preparation and procedural factors that affect results
Beyond the inherent limitations of breath testing, practical factors in how the test is prepared for and administered contribute substantially to result variability. Following the preparation protocol carefully is not optional. It is one of the few things you can control that directly affects the accuracy of your result.
- Antibiotics and antifungals should be discontinued at least 4 weeks before testing. Recent antibiotic use can suppress bacterial populations and produce false-negative results.
- Probiotics should be stopped at least 2 weeks before testing. Probiotic organisms can affect fermentation patterns and gas production.
- Prokinetic agents and laxatives should be held for at least 1 week, as altered motility changes the timing of substrate arrival in different gut segments.
- The preparatory diet (typically low-fiber, low-fermentation foods for 24 hours before the test) reduces baseline gas production so that any rise during the test is more clearly attributable to the substrate.
- Overnight fasting of at least 8 to 12 hours before the test establishes a baseline.
- Physical activity during the test should be avoided, as exercise can alter intestinal transit time and gas dynamics.
- Smoking and gum chewing should be avoided, as both can affect breath composition and swallowed air volume.
â ī¸If you did not follow the preparation protocol, tell your clinician before interpreting the results. A positive test after poor preparation may be a false positive, and a negative test after recent antibiotic use may be a false negative. The test may need to be repeated under proper conditions.
What helps with getting the most from your breath test
Given all these limitations, the practical question is how to maximize the usefulness of a test that is imperfect but often the best option available. The answer starts with preparation and extends to how the results are interpreted in context. Keeping a record of your symptoms, dietary patterns, and any medications or supplements you were taking in the weeks before testing gives your clinician the information they need to interpret the result properly. Tools like GLP1Gut can help you maintain this kind of log so that your test result is not evaluated in a vacuum but alongside a clear picture of your symptom pattern.
If you are choosing between glucose and lactulose substrates, discuss the options with your gastroenterologist. There is no universally correct choice. Glucose is generally preferred when proximal small bowel overgrowth is suspected, while lactulose may be more appropriate when distal overgrowth or methane production is the primary concern. Three-gas testing, where available, provides the most complete picture and should be preferred over two-gas testing when the option exists.
What is coming next in SIBO diagnostics
Several emerging technologies aim to address the limitations of current breath testing. Ingestible capsule devices equipped with gas sensors or sampling mechanisms could measure gases directly within the small intestine, eliminating the ambiguity about where the gas is being produced. Early prototypes have shown proof of concept, though none are currently available for routine clinical use (Kalantar-Zadeh et al., 2018).
Molecular diagnostics based on 16S ribosomal RNA sequencing or metagenomic analysis of jejunal aspirates could identify not just the quantity of bacteria but their species composition, potentially distinguishing pathogenic overgrowth patterns from benign shifts. This approach moves beyond the crude question of 'how many bacteria are there?' to the more clinically relevant question of 'which bacteria are there, and are they causing harm?'
Serum biomarkers, including volatile organic compounds measured in blood rather than breath, represent another avenue of investigation. These could potentially provide information similar to breath testing without the substrate preparation and timing challenges. However, standardization and validation of these approaches are in early stages, and clinical availability is likely still years away.
For now, the breath test remains the practical first-line diagnostic tool for SIBO. It is imperfect, but when performed correctly, interpreted in clinical context, and combined with a thorough symptom history, it provides useful information that can guide treatment decisions. The key is understanding what it can and cannot tell you, and not treating the result as the final word.
Is the lactulose or glucose breath test better for SIBO?
Neither is universally better. Glucose has higher specificity but only detects proximal small bowel overgrowth. Lactulose reaches the entire small intestine but produces more false positives because it inevitably enters the colon. Your gastroenterologist can recommend the more appropriate substrate based on your symptom pattern.
What is a three-gas breath test?
A three-gas breath test measures hydrogen, methane, and hydrogen sulfide. Traditional tests measured only hydrogen, or hydrogen and methane. Adding hydrogen sulfide detects an additional subset of patients whose overgrowth involves sulfate-reducing bacteria. The TrioSmart device is the first FDA-cleared three-gas testing system.
Can I do a breath test at home?
Yes. Several companies offer at-home breath test kits where you collect breath samples at timed intervals and mail them to a laboratory. The accuracy of home testing depends on how carefully you follow the preparation protocol and collection instructions. Results should always be reviewed by a qualified clinician.