Hot drinks and coffee sensitivity: why heat hurts differently from cold.

A tooth that objects to hot drinks needs the same two jobs done as one that objects to cold: the open dentine tubules narrowed, and the nerve behind them quietened, because heat and cold reach that nerve through the same channel. The caveat worth having before you buy anything is that almost every trial behind the ingredients in that aisle used cold water, cold air or a metal probe, and hardly any used heat¹². S3 Sensitivity Science™ pairs 5% potassium nitrate with two hydroxyapatites and full adult-strength fluoride in one daily paste, which you can check against the ingredient list, and it has no hot-drink result either.

What was checked25 peer-reviewed studies, S3 consumer trial (ADSL, 2026), NHS and Oral Health Foundation guidance

How hot does a tooth actually get when you drink coffee?

Less hot than the coffee, and by a wide margin. In 2022 a Korean group taped custom thermocouples to the outside of sixteen molars, eight of them intact and eight restored with gold inlays, and recorded the surface temperature in real time while people drank to a fixed regimen. On the intact natural teeth the maximum surface temperature during hot water was 40.5 °C, with a standard deviation of 1.2; the gold-restored teeth reached 44.7 °C, and their heating rate was about twice that of the natural ones³. On the cold half of the same clinical study, the intact teeth bottomed out at 31.5 °C and the restored ones at 25.0 °C³. Put those four numbers together and an intact tooth lives, across a hot drink and a cold one, inside a band of about nine degrees, while the drinks themselves span sixty. Mineral is a poor conductor, and the millimetres of it between a mouthful and a nerve do most of the work.

Three things that study is not. Those thermocouples were taped to the outside of the crown, so the record is of what the enamel reached and not of what the nerve underneath it experienced³. The drinks were hot water and cold water rather than coffee, their temperatures unstated, in sixteen molars belonging to people nobody describes as having sensitive teeth, with no pain scored at any point³. And the resting temperature was measured but is not in the published abstract, which is why this page will not tell you whether heat moves a tooth further from rest than cold does: the two ends are known and the middle is not³. It is the best in-mouth measurement there is, and it is small.

For what "hot" means in degrees, the coffee literature is more use than the dental one. A consumer study at the University of California, Davis, re-analysed more than 3,000 individual tastings of black coffee served across a range from 56 °C to 71 °C and found acceptance highest between 58 °C and 66 °C; only above 70 °C did most tasters call the coffee too hot⁸. A review of hot-beverage temperatures puts the brewing temperature of coffee and tea at around 93 °C, hot enough to scald, and recommends a service range of roughly 54 °C to 71 °C on the grounds that people prefer their drinks well below the temperature they are made at⁹. So the cup in front of you started at nearly boiling, is served twenty or thirty degrees below that, and lands on a tooth that will rise into the low forties.

What nobody has published is the recovery. The Seoul study reported the time each tooth took to reach its peak but not how long it took to return to rest, and no value for either is in the record³. If you have noticed that the second sip is easier than the first, that observation is yours and not the literature's.

Why does heat hurt differently from cold?

Because the fluid inside the tubule moves the other way, more slowly, and for longer. The mechanism everyone reasons from is the hydrodynamic theory, proposed by Brännström in the nineteen-sixties: pain from exposed dentine arises when fluid shifts inside the dentinal tubules and disturbs the nerve endings at their inner end¹⁰. Cold makes that fluid contract and draw outwards, away from the pulp. Heat makes it expand and press inwards. Same channel, opposite directions.

The only place those two directions have been quantified is a computer model. A 2018 fluid-structure-interaction simulation, built to dimensions measured by electron microscopy on eleven cats' teeth, put the peak fluid velocity under cold at about 410 micrometres per second outwards and under heat at about 205 inwards, with the modelled cold stress building roughly 2.9 times faster and the heat-induced stress lasting about 71% longer⁴. The greatest stress in the model fell on the tip of the odontoblast, the cell whose process sits in the tubule⁴. That is a simulation with feline geometry, not a measurement in a human mouth, and it earns a sentence about mechanism and nothing about how much anything hurts.

The nerve side has a matching asymmetry, and a matching caveat. Intradental nerves come in two classes, and animal recordings put numbers on both: fast A-delta fibres at a mean conduction velocity of 13.9 metres per second, with their receptive fields at the border between pulp and dentine, and slow C fibres at 1.3 metres per second, sitting deeper in the pulp¹¹. In the animal recordings that established the distinction, the fast fibres fired a burst only while the temperature was changing quickly and fell silent once it was steady, while the slow ones did not start until a mean latency of 7.3 seconds had passed and then ticked over at a low, regular rate¹¹. Those recordings were made by cooling cats' teeth, not heating them, and they are nearly forty years old¹¹. What they give a hot-drinks page is the vocabulary rather than the result: a sharp jab that arrives with the stimulus and leaves with it belongs to one system, and a dull ache that turns up seconds later and outstays its cause belongs to another.

Fluid movement is also no longer thought to be the whole story. A 2025 review of ion channels in dentine hypersensitivity describes a working model in which rapid fluid movement activates channels on the odontoblast and on the sensory neuron, which then release signalling molecules onto the nerve; the same authors conclude that these channels are poor drug targets because they do too many other jobs, and that occluding the tubule and desensitising the dentine remain the safest and most effective things to do¹². Which is convenient for a toothpaste page, and worth saying plainly: the current review of the fancier mechanism ends up recommending the two old ones. The Journal's page on why cold hurts your teeth takes the cold half and answers heat briefly in its FAQ; the resources page on what happens inside a tubule goes further into the plumbing, and the one on how desensitising toothpastes are tested explains the air blast and the probe.

Why do the trials almost never use heat?

Because the protocol the field agreed on did not include it, and nobody has revisited that. The 1997 consensus guidelines for the design and conduct of dentine hypersensitivity studies set out which stimuli a trial should use — tactile, cold and evaporative — along with a parallel-group design, blinding, and eight weeks for most of them¹. Heat is absent from that list, and the field has followed the list. A scoping review of 71 studies found the visual analogue scale and the Schiff cold-air scale to be the two measures the literature actually uses, often together¹³.

Follow that through the pooled reviews and the gap is complete. The Cochrane review of potassium toothpastes says its authors recorded sensitivity by thermal, tactile, air-blast and subjective methods across six randomised trials, and the results it publishes are the air-blast and tactile ones; its conclusion, on those, is that no clear evidence supports potassium toothpastes for dentine hypersensitivity⁵. The 2020 network meta-analysis of 125 randomised trials and 12,541 patients ranks formulations on tactile, cold and air stimuli¹⁴. The 2026 network meta-analysis of 93 randomised trials and 9,548 participants names its outcomes exactly: evaporative cold on the Schiff score, tactile on a Yeaple probe, and thermal cold². Where "thermal" appears in this literature, it means cold.

That claim is checkable, so here is how it was checked. Four searches were run on PubMed for this page in September 2026, and every title returned by the narrower three was read. Where "thermal" appeared in a desensitising-toothpaste study, it meant cold water or cold air every time.

The exceptions turned up on the last of those searches, and they are worth stating precisely rather than waving away. In 2010 a randomised parallel-design trial of 60 adults recruited for sensitivity to both hot and cold scored touch, heat and cold separately over 12 weeks; at 12 weeks the heat score had improved by 20.35% and 24.48% in the two test-tooth sets on the paste containing potassium citrate and zinc citrate, and by about 13% and about 17% on the comparison paste, while the cold scores improved by between 44% and 60% in every group⁶. There was no placebo arm and no significant difference between the two products, and neither active is one you will find on a UK sensitivity shelf⁶. Twenty-one years earlier, a six-week double-blind clinical trial of a dibasic sodium citrate dentifrice, which did not beat its control, provoked hypersensitive teeth with thermal, chemical and mechanical stimuli and reported that cold elicited the strongest response while heat and toothbrushing caused the least discomfort⁷.

Two trials, thirty-seven and sixteen years old, testing actives nobody buys for sensitivity, and neither of them appears in any of the pooled reviews above⁵¹⁴². So the accurate sentence is not that heat has never been tested. It is narrower and more useful: nothing on the UK sensitivity shelf has a hot-drink result and no pooled review reports one⁵¹⁴², and where heat was measured alongside cold it was the stimulus that moved least⁶⁷. Why heat was left out of the protocol in the first place is not something anyone has published. The guessing is easy and the evidence is absent, so this page stops at the guidelines.

Which ingredients have evidence, and what were they tested against?

All of them have evidence against cold, air or touch, and the honest way to read a shelf is to ask which stimulus produced the number on the box.

The most current ranking does not flatter the formula this page belongs to. The 2026 network meta-analysis concludes that stannous fluoride and arginine dentifrices should be considered first-line self-care options, with the choice guided by preference and tolerability rather than by an expectation that one beats another². S3 contains neither of those actives, which anyone can confirm from its ingredient list. In the same analysis, measured against a benchmark fluoride paste at two weeks on the cold-air score, nano-hydroxyapatite pastes showed a mean difference of −0.96 across two studies at moderate confidence, and potassium pastes with or without fluoride a mean difference of −0.42 at low confidence, described as a small but important effect². Its authors also note that most of the trials in the network were paid for by manufacturers².

For the two ingredients this page's reader is most likely to be choosing between, the picture by stimulus goes like this. Nano-hydroxyapatite has a 2019 systematic review and meta-analysis of six four-week randomised trials behind it, which found it ahead of its comparators on evaporative and tactile stimuli and found no difference on cold stimuli¹⁵. That exception is the one to hold on to here: cold liquid is the nearest thing in the literature to a hot liquid, and it is the stimulus on which the pooled hydroxyapatite advantage disappears. A double-blind randomised trial of 105 adults put a 15% nano-hydroxyapatite paste ahead of a fluoride paste and a placebo on cold-air and tactile sensitivity at two weeks and four¹⁶. An eight-week double-blind randomised trial funded by a nano-hydroxyapatite manufacturer, completed by 85 adults, used ice-cold water and an air blast and found pastes at 10% and 15%, with and without potassium nitrate, all reducing sensitivity from baseline and none of them differing from a calcium sodium phosphosilicate paste¹⁷.

Potassium nitrate is the older active, and its record is mixed in a way no packaging conveys. A four-week double-blind randomised trial of 30 adults found a 5% potassium nitrate paste ahead of a non-desensitising control on air-blast and cold-water sensitivity, while the companion electron-microscope test showed it had occluded no tubules at all, which is the mechanism working as designed rather than a failure¹⁸. Set against that, a 2019 network meta-analysis of 30 randomised trials found potassium toothpaste no different from placebo on its pooled outcomes, a 2017 network meta-analysis of nine trials found potassium nitrate reaching only a tendency towards relief, and the Cochrane review reached no clear evidence¹⁹²⁰⁵. A page that sells a potassium paste ought to print that paragraph, so it is printed.

Whatever the active, the clock is the same and it is measured against cold. In an industry-funded double-blind randomised trial of 120 adults, run by a company that makes a stannous fluoride paste, a marketed potassium nitrate paste beat a plain fluoride control on the cold-air Schiff score at day three and widened its lead to 44% by week eight²¹. Nothing in the desensitising literature acts on the first mouthful. Potassium accumulates around the nerve and mineral is laid down brush by brush, which is why relief is judged at two to four weeks rather than at the end of the first cup.

Where does S3 sit if hot drinks are the trigger?

On the mechanism, and this page will not pretend otherwise. Hydroxyapatite occludes and potassium desensitises, neither does the other's job, and a tooth that reacts to heat has both problems in it. S3 Sensitivity Science™ is built to that division of labour: potassium nitrate for the nerve, nano-hydroxyapatite inside the tubule, biomimetic hydroxyapatite on the surface, each doing something the others cannot. The percentages on the tube are inclusion levels of the ingredient as supplied rather than active content, and S3 states both, which is the kind of number worth checking on any paste you are comparing.

There is one figure from a hot drink, and it needs its full description. After one week of the S3 consumer trial, 81% of panellists said hot foods and drinks felt more comfortable. That work was commissioned by the brand, has not been published, and asked people how they felt instead of measuring a tooth, so it is a record of perception at one week and it sits at the bottom of the evidence hierarchy this page has used all the way down. Every study above it in this article can be looked up by anyone; that one cannot. Read it as what a panel of adults said after a week, and weigh it accordingly.

What S3 does not have, because no toothpaste has it, is a trial result against a hot stimulus²⁵. The honest chain runs like this. Heat and cold move the same fluid in the same tubule and arrive at the same nerve. The actives in the aisle have trial results against cold and touch². So the expectation for heat is carried across rather than demonstrated, and a reasonable expectation is not a measurement. The distance between those two things is the whole subject of this page.

When is hot-drink pain a dentist's question?

When the pain outlasts the drink. Dentine sensitivity is short and sharp and tied to its trigger: it arrives with the mouthful and leaves with it. An ache that begins a second or two after the hot liquid and is still there when the cup is empty is a different report, and it belongs in a surgery, not in a toothpaste aisle.

Three findings sit behind that advice. The first is that a hot tooth is hard to locate, even for the person who owns it. In a clinical study of 64 patients with symptomatic irreversible pulpitis, the exaggerated response to a thermal test was not confined to the inflamed tooth: the healthy tooth next to it, and the matching healthy tooth in the opposite jaw, over-responded too, and both settled once the inflamed tooth had been treated²². Every measurement in that study used a cold test rather than a hot one, so it is cited here for the spread and not for the trigger²². The second is that heat is a poor instrument even in trained hands: a 2018 systematic review and meta-analysis of 28 studies found the heat pulp test the least accurate of five ways of testing a pulp, with a pooled specificity of 0.67 against 0.84 for the cold test²³. The third is that the differential is genuinely wide. A 2019 clinical review written for UK primary dental care says a definitive diagnosis of dentine hypersensitivity is difficult, because the same reported symptoms have several possible causes²⁴. The same review puts a cracked tooth alongside it, notes that restorative work can inflame a pulp and raise a tooth's thermal sensitivity, and asks clinicians not to overlook the bite²⁴. If your hot-drink pain started after a filling or a crown, that last point is yours. Sensitivity that has arrived suddenly, in one tooth, is covered by the Journal's page on why are my teeth suddenly sensitive, and a tooth that reacts to hot and cold alike is a pattern of its own, taken by the sibling page on that combination.

The thresholds themselves come from the two bodies a UK reader should be using. The Oral Health Foundation names four triggers for booking: severe pain, a single affected tooth, a sudden onset, or sensitivity that has now run past a few weeks²⁵. Behind any of the four, it says, may be decay, a crack, a gum problem or an infection²⁵. The NHS puts its toothache line at two days, and brings it forward if painkillers are not touching the pain, if there is a high temperature, pain on biting, red gums or a bad taste, or if a cheek or jaw has swollen; it also says plainly that this is a dentist's job and not a GP's²⁶. Its advice while you wait is to keep away from anything very hot, very cold or sweet, which for this reader puts the coffee on hold and puts the toothpaste decision after the appointment²⁶.

Frequently asked questions

Will a sensitivity toothpaste such as S3 help with hot drinks?

Probably, and the evidence for it is indirect. No desensitising toothpaste on the UK shelf has been tested against a hot stimulus, and no pooled review of the category reports a heat outcome²⁵. What has been measured is the mechanism that heat and cold share: one open tubule, one fluid that moves when the temperature changes, one nerve at the far end⁴. Narrow the tubule and calm the nerve and you have addressed what a hot drink acts on, which is why the ingredients with cold and tactile trial results are still the ones to look for. S3's own hot-drinks figure comes from an unpublished consumer trial the brand commissioned, and records what a panel said after a week rather than what anyone measured in a clinic.

Why does my coffee hurt when ice cream does not?

Nobody has published an answer to that, and the honest response is a mechanism rather than a study. Heat and cold move the tubule fluid in opposite directions and at different speeds, so they are not the same provocation delivered two ways: in the simulation, cold produced the faster-building stress and heat the longer-lasting one⁴. The two nerve fibre classes also respond differently, one to rapid change and one after a delay of several seconds¹¹. If heat is your trigger and cold is not, describe that pattern to a dentist exactly as it happens: a UK clinical review notes that the same reported symptoms have several possible causes, and that restorative work can inflame a pulp and raise a tooth's thermal sensitivity²⁴.

Should I let my drinks cool down first?

It is the cheapest thing to try, and the temperature evidence is on its side even though nobody has run the experiment on a sensitive tooth. Coffee is brewed at around 93 °C and served far cooler; a consumer study of more than 3,000 tastings found acceptance highest between 58 °C and 66 °C, and only above 70 °C did most tasters call black coffee too hot⁹⁸. Most of the heat in a fresh cup, on those figures, is doing nothing you enjoy. Waiting a few minutes costs you nothing in flavour by that measure, and it lowers what the tooth is asked to absorb. Chasing a hot drink with a cold one is worth avoiding on the same reasoning, though no trial has run that sequence either.

My tooth aches for a minute after a hot drink. What does that mean?

It means book an appointment and describe the timing. Dentine sensitivity is sharp and brief and stops when the trigger goes; an ache that starts late and outstays the cup is a different pattern, and the differential a UK clinical review sets out for it includes a cracked tooth, an inflamed pulp and the aftermath of restorative work rather than exposed dentine alone²⁴. Do not try to work out which tooth it is on your own: when a pulp is inflamed, the healthy teeth beside it and the matching tooth in the other jaw can over-respond to a thermal test as well²². Even the clinic's own heat test is the least accurate of the five it can use, which is a reason to be examined rather than a reason to worry²³.

Does coffee damage teeth, or just stain them?

Staining is the familiar part and is not this page's subject. On erosion, the one measurement that separates a drink's temperature from its acidity is a laboratory study on bovine enamel in an artificial mouth, which found two fruit teas and one soft drink caused significantly more enamel loss when warm, while a cola, orange juice and citric acid did not change with temperature²⁷. Coffee itself was never put into that machine²⁷. Take from that only what it says: warming a drink can make some acids bite harder, on enamel in a machine, and which drink it is matters. The acid side of the question, including what coffee's own acidity does and how long to wait before brushing, belongs to the page on acidic food and drink triggers.