Skip to content
BeatSIBO

Reading a breath test result

By B. Gray Randolph · Updated August 5, 2026

If you are holding a result, this page is about what the numbers on it are being compared against, and how firm those comparisons are.

The published cutoffs

Hydrogen: a rise of 20 parts per million or more over baseline by 90 minutes. Methane: 10 parts per million or more at any point. Both from the North American Consensus. Notice they are different kinds of measurement, a rise within a window against an absolute level at any time, which is worth knowing before comparing them to each other.

Rezaie A et al., American Journal of Gastroenterology 2017 Clinical guideline

The gas with no threshold

Hydrogen sulfide has no criterion in that consensus. None. So a three gas test can give you a hydrogen sulfide value and there is no consensus cutoff to compare it against. The measurement is real; the interpretation is still being worked out at the level of the field.

We put this on the same page as the other two deliberately, because seeing the asymmetry in one place is the point. Two gases have agreed numbers behind them and one does not, and a report printing all three in the same format does not show you that.

Where thresholds come from

These are consensus figures. A group of specialists agreed them, which is a legitimate and normal way to produce a clinical cutoff and is not the same as a number derived from outcome data. That matters most at the margins: a value slightly under a consensus cutoff is slightly under a number that was itself agreed rather than discovered.

What is not settled here

There is an active argument that this testing should not be used for this purpose at all.

Kashyap P et al., Neurogastroenterology & Motility 2024 Peer-reviewed studyWe have not yet confirmed a direct link to this document, so we are not giving you one. The citation above is enough to find it, and we would rather say this than send you to a journal front page.

That position holds the methane and hydrogen sulfide criteria have never been validated. We are not able to settle that and we are not going to present the thresholds as firmer than the people who work with them do.

Why a positive is not a conclusion

Scalese G et al., Journal of Clinical Medicine 2025 Peer-reviewed studyWe have not yet confirmed a direct link to this document, so we are not giving you one. The citation above is enough to find it, and we would rather say this than send you to a journal front page.

Two things from that work belong next to any result. The two common substrates agreed on 8.7% of cases, so a positive on one is quite likely to sit alongside a negative on the other. And symptoms did not significantly differ between test-positive and test-negative groups, so the result cannot be corroborated by how you feel.

Pimentel M et al., American Journal of Gastroenterology 2020 Clinical guideline

And the guideline itself: all six recommendations conditional, five of six on evidence graded very low quality.

What to do with your result

Take it to a clinician with the substrate used, the timepoints, and any previous result. A number by itself is the least informative version of what you have. A number with its method, next to your history, is a conversation worth having.

Common questions

What counts as a positive hydrogen result?
A rise of 20 parts per million or more over baseline by 90 minutes, in the North American Consensus. That is a threshold agreed by a consensus body, which puts it among the better-established numbers in this area.
What about methane?
10 parts per million or more at any point during the test, in the same consensus. Note the difference in form: methane is an absolute level at any timepoint, hydrogen is a rise over baseline within a window. They are not the same kind of measurement.
And hydrogen sulfide?
There is no threshold. The North American Consensus contains no hydrogen sulfide criterion at all, so a test that measures it returns a number against a cutoff no consensus body has established.
Does a positive result mean I should be treated?
That is a clinical decision and it is not automatic. The guideline covering this rates five of its six recommendations as resting on very low quality evidence, and a positive on one substrate frequently comes with a negative on another.
My numbers were just under the cutoff. What does that mean?
That you were just under a threshold that was itself set by consensus rather than derived from an outcome study. A value near a cutoff is a reason for a conversation rather than a reason to conclude anything, in either direction.