Bad breath from the stomach? One case in a hundred

The sentence arrives with complete confidence, usually from someone who has already bought three mouthwashes. It's not my teeth. It's coming from my stomach. It is the most durable explanation for bad breath in circulation, and it survives because it feels anatomically obvious: the mouth opens onto a tube, the tube ends in a stomach, air comes out of the mouth. Something must be rising.

The verdict

Partly true — and the "partly" is very small.

In a Belgian multidisciplinary breath odour clinic that examined 491 patients with an ear-nose-throat specialist, a periodontologist and, when needed, a psychiatrist, oral causes accounted for 87 % of cases. The digestive tract accounted for 1 % ¹. That is one patient in a hundred, in a clinic that people reach precisely because everything simpler has already failed.

Where the smell is actually made

Bad breath is mostly one class of molecule: volatile sulfur compounds. Tonzetich, in the 1977 review that still anchors this field, showed that hydrogen sulphide and methyl mercaptan account for approximately 90 % of the total sulphur content of mouth air, and that the odour comes from putrefaction of sulphur-containing proteins by predominantly gram-negative oral microorganisms ². Brushing studies in that same work found plaque and tongue both to be important sources, with most of the odour emanating from the dorso-posterior surface of the tongue — the far back of it ².

That is worth sitting with, because it explains the whole misunderstanding. The back of the tongue is a fissured, papillated surface that holds a thick coat of shed epithelial cells, saliva sediment and food debris. Underneath the coat, oxygen runs out. Anaerobic bacteria then do what anaerobic bacteria do to protein: they strip the sulfur off cysteine and methionine and release it as gas ³. The substrate is your own dead cells. The conditions Tonzetich named for optimum putrefactive activity — low carbohydrate, physiological pH, anaerobic — are exactly the conditions on the tongue's back third when the mouth is closed and saliva is slow ².

The gums do the same thing in a periodontal pocket, which is why gum disease and bad breath travel together. In the Belgian series, tongue coating was behind 51 % of the oral cases, gingivitis 17 %, periodontitis 15 %, and combinations the remaining 17 % ¹. The organisms responsible are the familiar anaerobes of plaque and the gum line: when common oral anaerobes were incubated with cysteine and methionine and their headspace measured, Porphyromonas gingivalis was among the major producers of both hydrogen sulphide and methyl mercaptan ¹¹.

The share that really does not come from the mouth

The literature is consistent about the size of the non-oral slice, if not about its exact edges. Tangerman and Winkel put extra-oral halitosis of every kind at about 5–10 % of all cases ³. A 2022 systematic review of the aetiology literature put intra-oral factors at 80–90 % and extra-oral at 10–20 % ⁴. The Belgian clinic's non-oral 13 % broke down as ear-nose-throat problems 4 %, mixed ENT-and-oral 3 %, presumed psychiatric 5 %, and the digestive tract 1 % ¹.

Note what fills that non-oral slice in the Belgian series: ear-nose-throat problems and presumed psychiatric causes, each of them larger than the digestive tract ¹. And when a systemic condition does reach your breath, it does not do so by venting up the oesophagus. Odorous compounds circulating in the bloodstream cross into the lungs, volatilise into alveolar air, and leave with every exhalation ⁶. That is why blood-borne halitosis smells the same through the nose as through the mouth, while mouth-made halitosis does not. In a study of 58 people complaining of bad breath, 47 had an oral source, six had an extra-oral one, and five had no measurable malodour at all; all six extra-oral cases showed elevated dimethyl sulphide in both mouth and nose breath ⁵.

What the stomach bug Helicobacter pylori does and does not prove

The stomach's best claim is Helicobacter pylori. A 2016 meta-analysis of 21 studies found the odds of halitosis about 2.85 times higher in infected people than in uninfected ones, and a markedly lower rate of halitosis after successful eradication compared with failed eradication ⁷. Taken alone, that reads like a verdict for the stomach.

It is not, and the reason is a study that bothered to look at the mouths. A hundred dyspeptic patients with positive H. pylori serology were examined for bad breath, dental status by the DMFT index (decayed, missing, filled teeth) and oral hygiene by the simplified oral hygiene index. Halitosis tracked the dental index and the hygiene index, both significantly; the authors concluded they could not attribute halitosis in H. pylori patients to the organism with certainty ⁸. That same study found no significant relationship between halitosis and coated tongue in this particular group ⁸, so it is not a straightforward vote for the tongue either — but it does take the smell away from the stomach. The bacterium is also found in dental plaque and saliva as well as in the stomach; a review of the oral literature concluded that the mouth is its first extra-gastric reservoir ¹². An association with bad breath can therefore run through the mouth rather than through the gut. Eradication therapy is also a course of antibiotics, which does not leave the oral flora untouched. This is an association worth investigating in refractory cases, not a licence to skip the tongue.

What the myth costs the people who believe it

If you believe the smell comes from below, everything you do points downward: digestive enzymes, gastric remedies, elimination diets, endoscopies. Meanwhile the tongue goes unscraped. The Belgian team reported that many of their patients had already undergone diagnostic and therapeutic interventions to no avail before arriving ¹. In a separate inventory of 406 patients from the same clinic, most of those who returned the follow-up questionnaire were disappointed by the suggestion that their halitosis came down to insufficient oral hygiene ¹⁰. Disappointment is the honest human response to a boring solution — and the boring solution is the one with the evidence.

A systematic review identified five publications providing seven experiments comparing mechanical tongue cleaning plus toothbrushing against toothbrushing alone. Every experiment showed a positive effect on measures of oral malodour, though the reviewers were explicit that evidence for chronic halitosis specifically remains insufficient ⁹. That is a modest claim, and it is the one the data support.

So: clean the back of the tongue, get the gums assessed, and treat the mouth as the default suspect it statistically is. If the smell survives a genuinely clean mouth, the search moves outward — to the nose and sinuses first, then to blood-borne causes, and only then to the stomach. Mouth first, then nose, then blood, then stomach — that order is not a prejudice. It is the order the numbers were counted in.