Swallowing the saliva is normal and makes almost no difference to the nicotine you get; swallowing the pouch itself is a different question with a different answer. In the only study that measured it, under 5% of the nicotine released by a pouch was swallowed, and blood nicotine was no different whether people swallowed or spat. The nausea people blame on swallowing is a nicotine effect, not a digestion effect.
Two questions get asked as one here, and they come apart immediately. The first is about saliva: a portion is under your lip, it is producing more saliva than usual, and you are swallowing it. The second is about the object: the pouch has come loose, or a child has found the can. The evidence for each is completely different, and so is the answer.
One scope warning first, and it is important enough to put at the top rather than in a footnote: almost nothing in the literature below was measured in Swedish snus. The swallowed-fraction measurement was done on a tobacco-free pouch; the mass-balance figures come from American loose moist snuff, with subjects told not to swallow; the first-pass figure is for smokeless tobacco generally. I say which is which as I go.
What happens to nicotine you swallow instead of absorbing
Nicotine absorbed through the lining of your mouth goes straight into the bloodstream. Nicotine that goes down with your saliva takes a longer route: through the stomach and small intestine, then into the portal circulation and through the liver before it reaches the rest of you. The liver takes most of it on that first pass.
The EU’s own scientific committee on smokeless tobacco puts the number this way, attributing it to IARC: “Nearly 80% of nicotine that is absorbed from the intestine is metabolised to cotinine in the first pass through the liver and never reaches the systemic circulation.” The same opinion notes in passing that “Swallowing of the juice from STP is prevalent” and confirms the route: “whereas swallowed nicotine from STP will be absorbed from the small intestine.”
Why the liver gets so much of it is a matter of chemistry. As the standard pharmacology review explains, “Absorption of nicotine across biological membranes depends on pH. Nicotine is a weak base with a p”Ka of 8.0 — in acid conditions it is ionised and does not cross membranes readily. Your stomach is acidic, so swallowed nicotine sits there largely ionised and does most of its absorbing further down, which delivers it to the liver first.
This is not a snus phenomenon. It is also why nicotine gum delivers less than its label suggests: “The absolute dose of nicotine absorbed systemically from nicotine gum is much less than the nicotine content of the gum, in part, because considerable nicotine is swallowed with subsequent first-pass metabolism”. Across oral nicotine replacement products generally, “Bioavailability for these products with absorption mainly through the mucosa of the oral cavity and a considerable swallowed portion is about 50–80%”. Do not confuse those two figures, which look contradictory and are not: 50–80% is the whole product’s bioavailability, buccal and swallowed combined; the roughly 20% is the fate of the swallowed portion alone.
The figure European regulators actually work from is 44%. A 2024 risk assessment states: “Considering the oral human bioavailability of nicotine of 44 % of the applied dose and by application of an toxicological assessment factor of 10 for intraspecies variability, an acute reference dose (ARfD) of 0.8 µg/kg bw was derived by the EFSA in 2009” — the typo is in the original. That 44% is the number sitting underneath Europe’s acute reference dose for nicotine.
How much do you actually swallow? Somebody measured it
I expected to report that nobody had. That is no longer true, and the study is the most useful thing on this page.
A 2026 crossover trial in the Journal of Clinical Pharmacology set out “to explore whether nicotine absorption occurs solely through the buccal mucosa or if a significant proportion is swallowed”, in thirty-five habitual snus or pouch users. For one product arm, participants spat into a container every time they felt the urge to swallow, so the saliva that would have gone down could be measured.
The results: “This indicates that, based on the amount of nicotine in unused reference NPs (20.55 mg), 1.8% of the nicotine content of NPs may be swallowed during use for 30 min.” and, expressed against what the pouch actually gave up rather than what it contained, “on average 4.3% of the total amount of nicotine extracted from NPs during use may be swallowed”. Either way, “less than 5% of the total nicotine extracted from the NP was swallowed during 30 min of use, consistent with the nicotine PK curves.”
Then the direct spit-versus-swallow comparison. Peak concentration and total exposure were marginally higher when saliva was spat out rather than swallowed, and “the values were not significantly different (Table “S4) — the authors attribute the small gap to ordinary analytical variability. They also note the absence of the tell-tale sign of real swallowing: “also no noticeable secondary peak after 2–3 h following normal use that is typically observed when relatively large quantities of nicotine are swallowed”.
The funding has to be read alongside the numbers: “At the time the study was conducted, all authors were employees of BAT, which was the sponsor and funding source of this study.” The measurements are measurements; the conclusion the sponsor’s authors draw from them — “minimal ingestion by swallowing of nicotine occurs during NP use with little to no impact on the level and rate of nicotine absorption into the blood” — is their interpretation, and a tobacco company has an interest in that interpretation. The product was a 20 mg tobacco-free pouch used for 30 minutes. It was not snus.
For snus specifically the measurement does not exist. A BAT paper from 2013 lists it among the things it did not investigate: “various behavioral factors not investigated in this study could also affect nicotine absorption such as moving the snus portion around the mouth, spitting saliva during use, or swallowing some amount of loose snus”. The nearest adjacent figure comes from American loose moist snuff, where 24 subjects used 2 g for 40 minutes and were told to spit rather than swallow: “out of the 10 mg released over the use period, 75% was expectorated and only 25% absorbed in the BC.” That study was funded by Altria and its subjects were instructed not to swallow, so it tells you what leaves the mouth when you spit, not what happens when you do not.
If what you actually want to know is how to manage the saliva itself rather than what happens after it, that is the subject of snus and saliva: what the drip is.
So why does it make people feel sick?
Because of the nicotine, not because of the swallowing — and the mechanism is better understood in animals than in people.
Two areas of the brainstem lack a blood-brain barrier and therefore respond quickly to whatever is circulating: the area postrema and the nucleus of the solitary tract. A 2021 review of emetic mechanisms describes them as sites that “respond relatively rapidly to circulating and systemic stimuli in the blood and cerebrospinal fluid, which can trigger nausea and vomiting”. Acetylcholine is pro-emetic and “It activates the ligand-gated nicotinic receptor ion channels”. For nicotine specifically, the review’s evidence is that “Direct injection of nicotine can evoke vomiting in cats”, citing a 1987 experiment. That is the honest state of the mechanism: animal pharmacology, by direct injection, from the 1980s.
The human dose–response is weaker than you would expect. A pilot study gave eleven never-tobacco users six ascending nicotine doses in re-made pouches. Reported nausea climbed from essentially nothing at the zero dose to a mean of 12.0 at the top dose — but the statistics did not hold: “None of the DENS items were significant for a Dose X Time interaction or a main effect of Dose.” The only subjective item that moved significantly over time was excessive salivation. In a 2025 randomised pilot of pouches in smokers, “There were no severe adverse events. Two 6mg group participants reported nicotine toxicity symptoms, both involving dual use of cigarettes and pouches.” Both cases involved using cigarettes and pouches together, which is the mechanism worth remembering: stacking sources is what tips people over, not swallowing.
FDA’s own review of snus lists what excess nicotine does, and the list will be familiar to anyone who has gone too strong: “vertigo, headache, nausea, stomach ache, and heart burn.” Note that stomach ache and heartburn are on it. Swallowed saliva may well be why those two show up — the review of nicotine pharmacology notes that nicotine accumulates markedly in gastric juice through ion-trapping — but no study has demonstrated that causal chain, and I am not going to assert it.
There is one further downstream point that regulators do raise. FDA’s 2014 review of Swedish Match’s application noted that the “saliva produced during use of snus may be swallowed instead of expectorated, there is concern” that the nitrosamines in that saliva could raise the risk of oesophageal and stomach cancer. Read the modality exactly: FDA writes “there is concern”, not that this happens. Its conclusions were that for oesophageal cancer, “Although the risk of esophageal cancer associated with snus use is less than the risk in cigarette smokers, it is elevated over never-users of snus and nonsmokers.”, and that for stomach cancer the comparison with smokers is unclear. Germany’s risk assessment institute raises a related mechanism for pouches: that swallowed constituents can interact with nitrite in an acidic environment, so nitrosamines may form in the digestive tract itself.
Swallowing the pouch itself
A snus portion or a pouch is a non-woven fleece, typically viscose — regenerated cellulose — bonded with a thermoplastic, which is to say it is built much like a tea bag. It is not designed to be digested and it is not designed to dissolve.
A regulator has an actual case on file, and it is undramatic. Germany’s Federal Institute for Risk Assessment reports that “At the December 2020 session of the BfR Committee for the Assessment of Poisonings, representatives from the poison information centres in Germany reported on some cases of poisoning from nicotine pouches.” In one of them, “In one case, a pouch with 20 mg nicotine had been swallowed.” The outcome: “The affected person was given activated charcoal by the emergency service team but did not develop any symptoms other than stomach pain.” Of later cases, “Most of the symptoms reported involved nausea/vomiting and cold sweats.” The institute’s own instruction is one sentence: “The pouches should not be swallowed.”
That is the whole published record I could find on an adult swallowing one. A sealed pouch swallowed whole arrives in an acidic stomach, where nicotine is ionised and poorly absorbed, and is then metabolised on first pass — consistent with stomach pain and no systemic symptoms. It is still not a reason to do it: the fleece is not food, the dose is uncontrolled, and a pouch that comes apart releases its whole load at once rather than over half an hour.
Children, pets and accidental ingestion
This is the part of the question that actually matters, and the data exist.
A 2026 analysis in JAMA Network Open covered US poison-centre reports for children aged one month to five years, 2016 to 2023: 92,973 reported paediatric nicotine exposures, of which “99.6% were unintentional” and “Ingestion was the predominant route of exposure” — 80,054 of them. Within that, 6,525 ingestions (8.2%) involved the “other nontobacco nicotine” category, which is where pouches, gum and lozenges sit. On outcomes, the paper’s corrected text reads that 98.9% of exposures resulted in no or minor effects or were lost to follow-up, 1,023 in a moderate effect, 31 in a major effect, and two in deaths — both of those from e-cigarette liquid ingestion, neither from a pouch or snus. The authors note their own limitation: poison-centre data are “subject to underreporting and other systematic biases.”
A 2025 paper in Pediatrics narrows it to the product. I could only read its abstract — the publisher refuses automated access — so take this as the authors’ summary rather than a reading of the methods: across 134,663 nicotine ingestions in children under six, “The rate of nicotine pouch ingestions increased by 763.1% from 2020 to 2023.” and “Nicotine pouches were more likely to be associated with a serious medical outcome (odds ratio [OR]: 1.53, 95% CI: 1.10-2.13) or medical admission (OR: 2.03, 95% CI: 1.31-3.15) than other product formulations combined.” That is a rate rising off a small base, with confidence intervals that only just clear 1, and most ingestions overall had no effect. But a cheerful mint-flavoured can that looks like sweets is a genuinely bad object to leave on a coffee table, and the US lozenge carton says as much about its own product: “Nicotine lozenges may have enough nicotine to make children and pets sick.”
The “60 mg will kill you” figure is wrong, and a US government page still carries it
You will see it everywhere: 60 mg of nicotine is a fatal adult dose, so a few pouches would do it. It is an error with a traceable history.
A 2014 guest editorial in Archives of Toxicology traced it back. It opens: “Standard textbooks, databases, and safety sheets consistently state that the lethal dose for adults is 60 mg or less (30–60 mg)”, notes that this implies an oral LD50 far below what animal studies show, and follows the citation chain to nineteenth-century self-experiments described in an 1856 pharmacology textbook. Its revised estimate, with the author’s own hedge intact: “a careful estimate suggests that the lower limit causing fatal outcomes is 0.5–1 g of ingested nicotine, corresponding to an oral LD”50 of 6.5 to 13 mg/kg. Roughly ten times the textbook figure.
And the page that editorial criticised is still live. CDC’s NIOSH immediately-dangerous-to-life page for nicotine, last reviewed 4 December 2014 — fourteen months after the correction was published — states “The fatal human dose has been estimated to be about 50 to 60 mg [Lazutka et al. 1969].” and then, in the very next sentence, “An oral dose of 50 to 60 mg/kg is equivalent to” a particular inhalation exposure. Milligrams in one sentence, milligrams per kilogram in the next, about the same dose. That is a seventy-fold discrepancy inside one paragraph of a federal page.
None of which makes nicotine benign, and the classification actually in force says so. Europe’s risk assessment committee “proposed a classification of nicotine as Acute Tox. 2 (oral), with the hazard warning” “H300: Fatal if swallowed” and “with an acute toxicity estimate of 5 mg/kg bodyweight”, adopted into law in 2018. Mayer revised the lethal floor upward; he corrected an error, he did not issue a safety finding. Both things are true at once.
What the labels say, and the one that says nothing
Here is the asymmetry I find most telling: authorised oral nicotine products carry instructions about swallowing, and snus carries none.
The US nicotine lozenge — a product whose entire design is “dissolve it in your mouth” — instructs the user to “place the lozenge in your mouth and allow the lozenge to slowly dissolve. Minimize swallowing.” and, in bold on the carton, “Do not chew or swallow lozenge.” It caps use: “do not use more than 5 lozenges in 6 hours. Do not use more than 20 lozenges per day.” It warns that using them back to back “may cause you hiccups, heartburn, nausea or other side effects”. And for anything going wrong: “In case of overdose, get medical help or contact a Poison Control Center right away.”
Curiously, the nicotine gum label does not use the word “swallow” at all — zero occurrences in all 99,436 bytes of the Nicorette gum labelling I read. It manages the risk differently, by telling you what to do with the chewed piece: “nicotine gum may have enough nicotine to make children and pets sick.”
Now the snus can. Sweden’s food-authority regulation for snus requires exactly four things on a pack sold to a consumer: net quantity in grams, the date of manufacture, a storage instruction where shelf life depends on it, and the name and address of the manufacturer, packer or seller — plus the designation “snus” or “tuggtobak”. Searching the whole instrument for the Swedish words for swallow, ingest, directions and warning returns zero hits for each; the only “instruction” in it is a storage instruction. No authority anywhere instructs a snus user about swallowing. By contrast, a pouch sold through a European distributor catalogue does carry an instruction — one listing tells the user to “lay behind the upper lip for 5-30 minutes. Then spit it out without swallowing.” — but that is a company’s words, not a regulator’s.
One more thing in the Swedish file is hard not to enjoy. Livsmedelsverket switches off the drinking-water requirement for snus production on the ground that “produkter och tuggtobak förtärs inte” — these products are not consumed. That is the food authority’s own position, used to disapply a food rule, about a product whose users swallow its juices all day. I am not going to resolve that tension; both halves are quoted accurately and the contradiction is the authority’s, not mine.
The two questions, side by side
| Swallowing the saliva | Swallowing the pouch | |
|---|---|---|
| How common | Constant and normal; the EU committee calls it “prevalent” | Rare; a handful of poison-centre cases on record |
| How much nicotine is involved | Under 5% of what the pouch releases, measured | The whole pouch, uncontrolled |
| Effect on blood nicotine | No significant difference versus spitting | Not measured in any study I could find |
| Published symptoms | Nothing attributable to swallowing specifically | One case: stomach pain, no other symptoms, charcoal given |
| What a regulator says | Nothing, anywhere, to a snus user | “The pouches should not be swallowed.” |
| Main real risk | Nicotine dose from the product itself, not the route | Uncontrolled dose; and for a small child, the can rather than the pouch |
Frequently asked questions
Do I have to spit while using snus?
No, and the measurement says it makes no appreciable difference to the nicotine you get. FDA records the applicant’s own description of snus as products that “do not require expectoration during use” — note that this is the manufacturer’s characterisation, reproduced by FDA, and “does not require” is not the same as “spitless”.
Will swallowing the saliva make me feel sick?
The nausea is a nicotine effect and the dose you took is the variable that matters. Stomach ache and heartburn are both on FDA’s list of excess-nicotine effects, and swallowed nicotine is a plausible contributor, but no study has shown that chain. If a portion is making you queasy, the strength is the thing to change.
What if I swallow a pouch by accident?
The one documented adult case produced stomach pain and nothing else. Nicotine swallowed in a sealed pouch is poorly absorbed in an acidic stomach and heavily metabolised on first pass. That is not advice about what to do — if you are worried about a swallowed pouch, particularly for a child, a poison control centre is who to call, and the lozenge label’s own instruction is to get medical help right away in case of overdose.
Does swallowing waste the nicotine?
Mostly — roughly 80% of nicotine absorbed from the intestine never reaches the systemic circulation. But since under 5% of what a pouch releases is swallowed, the loss is too small to show up in blood.
Is swallowing more of a problem with snus than with pouches?
Possibly, and for a reason nothing to do with nicotine: snus saliva carries nitrosamines, which a tobacco-free pouch does not. FDA raised exactly that concern in 2014 and did not resolve it, and its 2023 renewal review does not return to the subject. See TSNAs in snus and nicotine pouches vs snus.
Sources
- SCENIHR, Health Effects of Smokeless Tobacco Products, adopted 6 February 2008 — first-pass metabolism of swallowed nicotine, the route, and the prevalence of swallowing. The 80% figure is attributed there to IARC 2007.
- Benowitz NL, Hukkanen J, Jacob P 3rd, “Nicotine Chemistry, Metabolism, Kinetics and Biomarkers”, PMC2953858 — pH and ionisation, gum bioavailability, the 50–80% range for oral NRT, gastric ion-trapping.
- Reimann H et al., “Beyond smoking: Risk assessment of nicotine in pouches”, Toxicology Reports 13 (2024), PMC11568777 — the 44% oral bioavailability figure and EFSA’s 2009 acute reference dose. Authors declare no competing interests.
- Azzopardi D et al., “A Randomized Crossover Clinical Study to Assess the Effect of Oral Nicotine Pouches Used for Different Durations on Plasma Nicotine Pharmacokinetics”, J Clin Pharmacol 66(1):e70090, 2026 — PMC12709140. Sponsored and funded by BAT; all authors were BAT employees at the time. The swallowed-fraction measurement and the spit-versus-swallow comparison.
- Digard H et al., “Determination of Nicotine Absorption from Multiple Tobacco Products and Nicotine Gum”, PMC3524070 — swallowing listed among factors not investigated. BAT R&D authors. Rostami AA et al., Scientific Reports 2022, PMC8776883 — the moist-snuff mass balance. Funded by Altria Client Services.
- Zhong W et al., “Mechanisms of Nausea and Vomiting”, Int J Mol Sci 2021, PMC8198651 — area postrema, nicotinic receptors, and the 1987 cat experiment it cites. NIH-funded, no declared conflict.
- Ozga JE et al., “Acute Effects of Snus in Never-Tobacco Users: A Pilot Study”, PMC6097526 — the dose series and the non-significant dose effects; university and NIGMS training-grant funded. Fucito LM et al., PMC12377922 — NIDA- and FDA-funded; the two toxicity-symptom cases.
- FDA, Technical Project Lead review of Swedish Match’s modified risk tobacco product applications (2014) — excess-nicotine effects, the swallowed-nitrosamine concern, and the oesophageal and stomach cancer conclusions. FDA’s 2023 renewal review — the applicant’s “do not require expectoration” description.
- BfR, Health risk assessment of nicotine pouches, updated opinion 023/2022, 7 October 2022 — the swallowed-pouch case, the instruction not to swallow, the ARfD, the ECHA classification, and endogenous nitrosamine formation.
- Rosen PE et al., “Pediatric Nicotine Exposures Reported to US Poison Centers”, JAMA Netw Open 2026;9(3):e260479 — PMC12961511. Corrected 22 May 2026; the outcome figures quoted are the corrected ones. No funding statement; no conflicts reported.
- Olivas M et al., “Nicotine Ingestions Among Young Children: 2010-2023”, Pediatrics 2025;156(2):e2024070522, doi 10.1542/peds.2024-070522 — abstract only; the publisher refuses automated access, so the methods and funding were not read.
- Mayer B, “How much nicotine kills a human?”, Arch Toxicol 88(1):5–7, 2014 — PMC3880486. CDC/NIOSH, Nicotine IDLH documentation, last reviewed 4 December 2014.
- Nicorette nicotine polacrilex lozenge label and nicotine gum label, DailyMed — read 6 November 2026.
- LIVSFS 2024:2, 12 § and 13 § — what a snus pack must carry, and what it need not. Kontrollwiki article 887 — the “not consumed” position.
- Last verified 6 November 2026.
