Why whitening makes teeth sensitive: peroxide, the pulp, and why it is not ordinary sensitivity.

Whitening makes teeth sensitive because the peroxide does not stay on the surface: it crosses the enamel and dentine to the pulp, the living tissue at the centre of the tooth, and the leading explanation, built mostly on laboratory and animal studies, is that it sets off inflammation and a chemical-sensing nerve channel there¹². That differs from the fluid movement in exposed dentine behind ordinary sensitive teeth, which is why bleaching pain can arrive in a healthy tooth with nothing touching it³. Human teeth examined after high-strength in-office bleaching do show inflammation in the pulp, but no study has tied those changes to what the person felt, and a 2015 systematic review of 18 trials found no evidence-based explanation for bleaching pain and called it an unsolved phenomenon⁴⁵. In S3 Sensitivity Science™, potassium nitrate is there to calm the nerve, nano-hydroxyapatite to work inside the dentine tubule and biomimetic hydroxyapatite to work on the surface; the leading explanation puts bleaching pain deeper, in the pulp, and no trial has tested S3 against it. The pain is usually brief: in a randomised trial of 120 adults after in-office bleaching, it was worst in the first 48 hours and close to none by the seventh day⁶.

What was checked32 peer-reviewed papers, the NHS and the Oral Health Foundation

Why does whitening make teeth hurt at all?

Because peroxide has to get inside a tooth to change its colour, and the inside of a tooth is alive. Hydrogen peroxide, and carbamide peroxide, which releases it, can penetrate the hard tissues of the tooth because the molecule is so small, moving through the enamel into the dentine and on towards the pulp, the soft core of nerves and blood vessels¹. When this page says "the nerve", it means the nerve fibres in that pulp. The dentine is threaded with microscopic tubules that run towards the pulp, and on the leading explanation of bleaching pain they are the road the peroxide travels, not the trigger. The Journal's page on what dentine tubules are shows how they are laid out.

It is common, and how common depends on the gel, the technique and how people are asked. At lower strengths, over a few weeks, the figure sits around half: in a clinical study in which 172 people recorded their symptoms every day through a fortnight of bleaching, 47% had some sensitivity¹². In a double-blind randomised trial of 100 adults using a 15% carbamide peroxide gel at home for four weeks, 54% reported mild sensitivity¹¹. With high-strength in-office gels it is usually most people: in a double-blind randomised trial of 140 adults, 74% of those given 35% hydrogen peroxide reported sensitivity¹³. The authors of a multicentre randomised trial of 90 adults treated with 35% hydrogen peroxide wrote that most patients get it, whichever arch is bleached¹⁴. No single figure describes everyone.

The NHS lists teeth becoming sensitive to cold or sweet food and drink among the side effects of whitening done by a dental professional, beside sore gums or throat and white patches on the gums, and tells you to talk to your dentist if one of them bothers you¹⁵. The Oral Health Foundation describes temporary sensitivity, especially to cold, and mild gum irritation, and calls both usually short-term¹⁶.

What happens once the peroxide arrives is the part nobody has watched directly. The leading explanation, pieced together from extracted teeth, cell cultures, rats and a small number of human teeth removed after bleaching, is that peroxide reaching the pulp causes oxidative stress and inflammation, and that it switches on TRPA1, a channel on pain-sensing nerve fibres that responds to oxidising chemicals³¹⁰⁴.

How does peroxide get from the surface of the tooth to the pulp?

By diffusing through the enamel and dentine, which laboratory studies on extracted teeth have measured directly¹⁸. In a laboratory study of 36 extracted human teeth kept under a simulated pulpal pressure, hydrogen peroxide reached the pulp chamber from both a 6% paint-on varnish and a 40% gel, with the largest amounts in molars and significantly more from the 40% gel in premolars and molars¹. Some of the concentrations found in the chamber reached a level reported as toxic to pulp cells in culture, none passed the higher benchmark the authors cite, and they point out that a living pulp has defences against free radicals¹.

Repeating the application pushes more through. In a laboratory study on enamel-and-dentine discs cut from extracted third molars, peroxide from each of five commercial bleaching products crossed into an artificial pulp chamber, the amount rose with every application, and three 10-minute applications delivered more than one 30-minute application⁸. Shining a light on the gel significantly increased diffusion for three of the five products, and the products' stated concentrations did not predict how much got through⁸. The authors add a caution that applies to every laboratory result in this section: an extracted tooth has no pulpal pressure and no outward flow of fluid, so the amounts may be higher than in a living mouth⁸.

Strips work the same way. In a laboratory study of 24 extracted upper front teeth, more peroxide reached the pulp chamber from a whitening strip containing 14% hydrogen peroxide than from one containing 6.5%⁷. In a 2026 laboratory study of 96 extracted premolars, two over-the-counter peroxide strips also delivered hydrogen peroxide into the pulp chamber, in smaller amounts than a 6% hydrogen peroxide home gel, and a peroxide-free strip delivered none¹⁷.

So does the amount of gel. In a laboratory study on bovine tooth discs, 120 microlitres of 35% hydrogen peroxide gel pushed more peroxide into an artificial pulp chamber than 30 or 60 microlitres, without whitening the discs any further, and in the animal arm of the same study larger volumes on rat molars caused more pulp inflammation and oxidative stress a day later⁹.

Only one study read for this page measured the pulp of teeth that stayed in the mouth¹⁸. In a randomised trial of 80 adults, a pulse oximeter on the upper front teeth recorded mean pulp oxygen saturation falling by two to five percentage points in all four groups straight after an in-office session with 35% hydrogen peroxide, significantly in one group, and sitting above its starting level a month after the course¹⁸. The authors explain the dip as inflammation inside the rigid pulp chamber squeezing its blood supply¹⁸. It shows the pulp registering the procedure, and it is not a measure of pain: the trial did not analyse pain against it, although 2.5% of participants were withdrawn because of severe tooth sensitivity¹⁸.

What happens in the pulp when the peroxide arrives?

In human teeth removed after bleaching, inflammation, and after high-strength gels in lower incisors, small areas of dead tissue; in cells and rats, oxidative stress and a rise in a pain-sensing channel. Each finding rests on a different kind of evidence, and none has been tied to what a person feels.

What human teeth removed after bleaching show

Some people who needed teeth taken out, often premolars or third molars removed for orthodontic treatment, have agreed to have those teeth bleached first so that the pulp could be examined under a microscope. A 2024 systematic review gathered 30 in vivo studies of this kind, in people and in animals: of the seven studies on human teeth that looked at inflammation, five found significant inflammation in the pulp of bleached teeth, and in eleven of the twelve studies that looked for it, new tertiary dentine was laid down on the pulp side after 35% to 38% hydrogen peroxide⁴. The review judged the human studies at low risk of bias, could not pool them, and concluded that high-concentration gels increase pulp inflammation and necrosis shortly after bleaching, mainly in rat molars and human incisors⁴.

The individual human studies sort by the strength of the gel and by the kind of tooth. They are small, three of them used teenagers' teeth, and none could follow a bleached tooth over time, because each tooth had to come out to be examined.

Read down the table, the low-strength home gel produced slight changes that were no longer significantly different from untreated teeth two weeks after bleaching stopped¹⁹. The same home gel produced no measurable rise in two pain-related neuropeptides²⁰. High-strength in-office gel produced more inflammation than a home gel in a randomised study of third molars²¹. In lower incisors examined two days after high-strength gels, it left areas of necrosis in the pulp of the crown, in a clinical study and a randomised one²³²⁵. Premolars given a 38% gel looked normal or nearly so in that clinical study and in a randomised study whose funding is not stated²³²². The two small studies that also recorded sensitivity found it highest in the groups whose incisors showed necrosis²⁵²⁴. That is a pattern across groups of five or fewer teeth, not proof that the necrosis is the pain.

What this evidence cannot show matters as much. None could watch a bleached tooth for years, because each tooth was removed, and the review asks for histology with long-term follow-up⁴. The one trial that kept measuring living incisors found their pulp oxygen saturation above its starting level a month after the course¹⁸. And the teeth were often young: the authors of the 2023 randomised incisor study name their participants' youth as a limitation²⁵.

What cells and rat teeth show

The chemical side of the story comes from models. A 2010 hypothesis paper noted that bleaching pain can arise in healthy, intact teeth without any trigger, which does not fit a mechanism that needs a stimulus on exposed dentine, and proposed instead that hydrogen peroxide switches on TRPA1, a channel carried by the pulp's sensory nerve fibres that opens in response to oxidising chemicals³. It was a proposal, not a finding: the paper collected no data.

Two later papers from one university group in China tested the idea in a model. Dental pulp stem cells were grown on the dentine side of enamel-and-dentine discs cut from extracted premolars, and bleaching gel was painted on the enamel side. In that laboratory set-up the gel raised oxidative stress, calcium and ATP inside the cells and switched up TRPA1 and the inflammatory signals TNF-alpha and IL-6, more with a 40% gel than a 15% gel, and a TRPA1 blocker or an antioxidant damped the response²¹⁰. In rat incisors treated with the same gels, TRPA1 and a related channel protein, PANX1, rose in the pulp²¹⁰. The authors read this as TRPA1 playing a critical role in bleaching sensitivity; what they measured, in the laboratory and in rats, was cell chemistry and tissue proteins².

A rat study adds a time course. After a 38% hydrogen peroxide gel, the coronal pulp of rat molars showed necrosis and inflammation straight away, and both the inflammation and the staining for substance P and CGRP, two neuropeptides involved in pain signalling, fell over the next two days²⁷. In a clinical study of people given a 10% carbamide peroxide home gel, those two neuropeptides did not rise measurably in premolar pulp²⁰. In a clinical study of premolars removed after a single in-office application, another neuropeptide linked to neurogenic inflammation, neuropeptide Y, was higher, significantly only after a 40% gel²⁶.

Has the channel itself been measured in a person? Not in anything a search run for this page could find; the terms and what came back are in the table at the end of this section. TRPA1 has been measured after bleaching in cells and rat teeth, not in a living human tooth²¹⁰.

Markowitz's hypothesis made a prediction worth checking: if the channel is the cause, treatments that damp nerve excitability, such as potassium salts, should be the treatment of choice³. The trials of that idea have not borne it out strongly: a 2021 systematic review and meta-analysis of 24 studies found that potassium nitrate applied before bleaching lowered the risk by 12%, a result the reviewers themselves describe as "subtle and clinically questionable"²⁸. The page on potassium nitrate before bleaching reads those trials one by one.

What the clinical trials say about the mechanism

Set beside the models, the clinical literature is blunt. When reviewers went looking for the cause in 2015, their systematic review of 18 randomised trials turned up no evidence-based explanation at all, on top of trials that were mostly at high risk of bias and that scored severity and duration in ways that could not be compared; the title they chose calls bleaching sensitivity an unsolved problem⁵. This page does not claim to have solved it.

Is it the same pain as ordinary sensitive teeth?

No, on the leading explanations of each, and the difference accounts for most of what people notice. Ordinary sensitive teeth, dentine hypersensitivity, give a quick, sharp pain when something hot, cold, sweet or acidic reaches dentine that has lost its cover, very often where gums have shrunk back or enamel has worn²⁹. The hydrodynamic theory explains it as fluid moving in open tubules and stimulating nerve endings, and the page on the hydrodynamic theory sets out that account and its limits; the page on what happens inside a dentine tubule when you drink something cold follows one stimulus in. Bleaching pain can come in a tooth with no exposed dentine and no stimulus, and on the leading explanation it starts in the pulp³.

How often bleaching pain comes with no trigger at all is not known: the 2010 hypothesis paper describes it, but neither the clinical study that kept daily diaries through a fortnight of bleaching nor the 2015 systematic review of trials reports it as a separate measure in its abstract³¹²⁵.

The practical consequence runs through every page on whitening here: what works for everyday sensitive teeth cannot be assumed to work for bleaching pain, and what helps bleaching pain says nothing about everyday sensitivity. People who already have sensitive teeth can have both at once, and in a double-blind randomised trial of 100 adults using a home gel, reported sensitivity correlated with gum recession¹¹. That is an association, not a mechanism.

Where do the actives in S3 act, and is that where bleaching pain starts?

Not where the leading explanation puts bleaching pain. S3 was designed around ordinary sensitive teeth, in which a tooth has two things wrong at once, a nerve that fires too readily and dentine tubules that are open. Its potassium nitrate is there for the nerve, its nano-hydroxyapatite for the inside of the tubule and its biomimetic hydroxyapatite for the surface.

Bleaching pain, on the evidence above, starts somewhere else and for another reason. By the time the pulp responds, the peroxide has already crossed the enamel and dentine, the tooth may have no exposed dentine at all, and the change seen in human teeth is inflammation inside the pulp itself⁴. Even for ordinary sensitive teeth, the account of how potassium quietens nerve fibres through the tubules had not been confirmed in intact human teeth when a 2000 review of the potassium trials looked³².

So the answer has three parts. No trial has put this toothpaste into a bleaching course, before, during or after. Nothing on this page suggests that its actives reach the pulp or act on bleaching pain. And the actives it shares with the bleaching literature were tried there in other products and other forms, such as gels worn in trays, pastes brushed on before appointments and minerals mixed into the bleaching gel itself, with mixed results; the page on nano-hydroxyapatite and bleaching sensitivity goes through the mineral work one paper at a time.

Why does the strength of the gel matter, and does the light?

Because strength, contact time and volume set how much peroxide reaches the pulp, and in the trials the pain follows the dose. The laboratory work shows the dose travelling: more from a 40% gel than a 6% varnish in back teeth¹, more from 14% strips than 6.5% strips⁷, more with every reapplication⁸ and more with a bigger dollop of gel⁹.

In people, the largest trial here is clear. In a double-blind randomised trial of 140 adults, 44% had sensitivity with 6% hydrogen peroxide against 74% with 35%, and the stronger gel whitened more¹³; the hub page on how to whiten sensitive teeth without making them hurt weighs that trade. A smaller randomised trial of 87 patients, whose funding is not stated, points the same way less cleanly: sensitivity straight after in-office bleaching scored 24 out of 100 with a light-activated 15% hydrogen peroxide gel against 42 with a light-activated 35% gel, but the weaker gel came with a catalyst added, so strength is not the only difference³⁰.

The pulp studies point the same way: in a randomised study of third molars examined a week after bleaching, in-office 38% hydrogen peroxide produced more inflammation than a 15% carbamide peroxide home gel²¹. How the two routes compare for sensitivity is contested, and the page on in-chair whitening versus home trays weighs the evidence.

Light is the odd one out. In the laboratory, shining a light on the gel pushed significantly more peroxide through for three of five products⁸. In the randomised trial above, whose funding is not stated, the light did not change how much it hurt³⁰. The NHS describes laser whitening at a dental surgery as quicker and more effective than gel alone¹⁵, and the page on in-chair versus home trays sets that view against the reviews.

One more consequence of an inflammatory explanation needs saying, because it sounds like an obvious fix. A 2019 systematic review and meta-analysis of 11 randomised trials found that anti-inflammatory and pain-relief drugs given to prevent in-office bleaching sensitivity did not lower its risk or intensity, at high certainty³¹. Which painkiller to take, if any, is a question for a pharmacist or the dentist.

If you are booking, the useful questions are about dose rather than products: what strength of peroxide the dentist plans to use, for how long and how many times a session, whether a lower strength or shorter wear is an option for you, and what they would like you to do if it hurts.

How long does it last, and who gets it worst?

Usually days, and most intensely in the first one or two. In a single-blind randomised trial of 120 adults after in-office bleaching with 40% hydrogen peroxide, sensitivity was highest within the first 48 hours and approached zero by the seventh day in every group⁶. In a clinical study in which 172 people kept a daily record through two weeks of bleaching, 77% of everyone taking part reported three or fewer days of sensitivity, temperature sensitivity tended to come later in the fortnight, and for some people it went on much longer¹². In the placebo-controlled trial of a 15% carbamide peroxide home gel, nobody still had severe sensitivity by the second week, and nobody had moderate sensitivity by the fourth¹¹.

Some of it would have happened anyway, which is worth knowing and is no reason to doubt anyone's pain. In that double-blind randomised trial of 100 adults, 54% reported mild sensitivity on the placebo gel, the same share as on the real gel; moderate sensitivity was reported by 10% on the real gel and 2% on placebo, and severe sensitivity by 4% on the real gel and nobody on placebo¹¹. In a double-blind randomised trial of 54 adults whose funding is not stated, 12% on a 9.5% hydrogen peroxide strip and 11% on a placebo strip reported minor sensitivity³³. Mild twinges were common with or without peroxide; in the gel trial, the stronger pain came with the peroxide.

Lower teeth come off worse in the one trial designed to check. In a multicentre randomised trial of 90 adults who had both arches bleached in-office with 35% hydrogen peroxide, the risk and the intensity of sensitivity were higher in the lower arch, while gum irritation and colour change did not differ¹⁴. The trial does not say why. In the human pulp studies above it was lower incisors that showed necrosis after high-strength gel, but no one has tested whether that explains the lower arch²³²⁵. In the 87-patient trial whose funding is not stated, tooth thickness measured by CT did not predict sensitivity³⁰.

The long view is less tidy. Four and a half years after a split-mouth randomised trial of in-office 35% hydrogen peroxide that had enrolled 62 patients, 25.5% of participants reported long-term tooth sensitivity³⁴. Everyone in that study had been bleached, and there was no unbleached group, so nobody can say how much of that sensitivity the bleaching caused; nobody should say it caused none either.

The sources behind these figures, and what each compared:

Is the enamel being damaged?

Not in a way that explains the pain. In laboratory studies, peroxide bleaching reduces enamel microhardness by a small amount: a 2026 systematic review and meta-analysis of laboratory studies, nine pooled from 81 eligible, put hardness after bleaching at 0.89 of hardness before and found no clear evidence of a clinically meaningful effect³⁵. In the laboratory study of over-the-counter strips, the enamel surface looked intact under the electron microscope in every group¹⁷.

The popular picture, in which whitening strips the enamel off and leaves the tooth exposed, is not what the mechanism research proposes. The leading explanation needs the enamel to stay where it is and let the peroxide through, which is how peroxide ends up in the pulp chamber of an intact extracted tooth¹. The pain is therefore not a sign that enamel has gone. It is not a sign that nothing has happened either: the human pulp studies above found inflammation after high-strength gels⁴. For the practical side of the days after a course, see the Journal's page on sensitive teeth after whitening.

When is it not bleaching sensitivity?

When it behaves like toothache rather than fading twinges. Bleaching sensitivity, as described above, is worst early and fades over days. A pattern that points somewhere else is one tooth hurting far more than the rest, pain that carries on long after a cold drink or builds instead of easing, pain that wakes you, pain when you bite, or any swelling.

The NHS advises booking a dentist for toothache that lasts more than two days, that painkillers do not settle, or that comes with a high temperature, pain on biting, red gums, a bad taste or a swollen cheek or jaw, and sends people to A&E if swelling spreads around the eye or neck or makes breathing, swallowing or speaking difficult³⁶. For sensitivity in general, the Oral Health Foundation's reasons to get checked are severe pain, sensitivity that goes on for more than a few weeks, a single affected tooth and pain that starts suddenly, any of which can point to decay, a crack, gum problems or infection²⁹. After a whitening course, the dentist who treated you is the first call, and the NHS's whitening page says to talk to your dentist about any side effect that bothers you¹⁵.

A tooth that is clearly worse than all the others, the night after an appointment, is worth a call to the practice the next morning, and so is sensitivity that is getting worse instead of better.

Frequently asked questions

Why do my teeth zing after whitening when nothing is touching them?

Because, on the leading explanation, the trigger is already inside the tooth. Peroxide from the gel crosses the enamel and dentine and reaches the pulp, and laboratory and rat studies suggest it provokes inflammation and a chemical-sensing channel there, so the pulp's nerve fibres can fire without a cold drink or a toothbrush¹². Ordinary sensitive teeth, by contrast, hurt when something reaches exposed dentine. Nobody has recorded the channel at work inside a living human tooth, and a systematic review of the bleaching trials called the phenomenon unsolved⁵.

Does whitening sensitivity mean my enamel is damaged?

No. The peroxide passes through enamel that stays in place, and in laboratory studies the loss of enamel hardness after bleaching is small, with no clear evidence of a clinically meaningful effect in a 2026 systematic review and meta-analysis³⁵. The pulp beneath is a different matter: studies of human teeth removed after high-strength bleaching found inflammation there⁴.

How long does sensitivity last after whitening?

Usually a few days. After in-office bleaching, in a randomised trial of 120 adults, it peaked within 48 hours and was close to zero a week later⁶. A trial that followed patients for four and a half years found a quarter reporting tooth sensitivity, with no unbleached group to show whether bleaching was the cause³⁴. If it is getting worse, or one tooth stands out, speak to the dentist who treated you.

Is bleaching sensitivity the same as the sensitivity S3 is made for?

No, on the leading explanation of each. S3 is made for dentine hypersensitivity, with potassium nitrate for the nerve and two forms of hydroxyapatite for the inside of the tubule and the tooth's surface. Bleaching pain is thought to begin in the pulp after peroxide has crossed the tooth, no trial has tested this toothpaste in anyone having their teeth bleached, and this page does not suggest it for bleaching pain.

Why do my bottom teeth hurt more?

In one multicentre randomised trial of 90 adults given in-office bleaching on both arches, the risk and intensity of sensitivity were higher in the lower arch¹⁴. Why is not known. Lower incisors examined after high-strength gels in two small human studies showed areas of necrosis in the pulp, but no study has linked that to the lower arch, and tooth thickness did not predict sensitivity in a trial that measured it by CT, whose funding is not stated²³²⁵³⁰.