Whitening toothpastes and sensitivity: abrasivity, RDA and what to avoid.

A whitening toothpaste lifts stain off the surface of the teeth, by scrubbing or with a chemical cleaner, or tints that surface with a blue pigment, and in the UK it may contain or release no more than 0.1% hydrogen peroxide, so it cannot bleach a tooth the way a dentist's gel does¹. The case that whitening toothpastes make sensitive teeth worse is thinner than its reputation: most laboratory studies find them no harder on enamel than regular pastes or brushing with saliva²³, and almost nobody has tested one on people recruited because their teeth were sensitive. S3 Sensitivity Science™, whose ingredient list on its own product page relies on hydrated silica and names no peroxide and no blue pigment, is in the same position as almost every tube in this site's September 2026 scan of the UK shelf: it publishes no abrasivity figure.

What was checked27 peer-reviewed studies and reviews, UK legislation, an OPSS product safety report, CAP advice and an ASA ruling, the NHS, the Oral Health Foundation, the European Commission's CosIng database, and product information as published by each brand

What does a whitening toothpaste actually do to a tooth?

It works on the outside of the tooth, in one of four ways, and only one of them uses a bleaching chemical, which UK law keeps to a trace¹. Strips, pens and gels are a separate subject, covered on the page on whitening strips and sensitive teeth.

The first way is to take stain off. Abrasives such as hydrated silica, perlite, calcium carbonate, sodium bicarbonate and calcium pyrophosphate polish the surface film away, and some pastes add a chemical anti-stain agent such as a pyrophosphate or pentasodium triphosphate. A 2018 systematic review and meta-analysis of 21 publications, whose funding is not stated in its record, found that whitening toothpastes reduced both the area and the intensity of natural surface stain more than regular toothpastes, with a stain-area difference of −0.44 on the original Lobene index, and a similar pattern with or without an added chemical agent⁵. An updated 2025 systematic review pooled 14 randomised trials and reached the same conclusion, although the trials differed from one another so much that heterogeneity reached 91%⁶. The Oral Health Foundation says something close about shop-bought strips, gels and paint-on products: they may help remove some surface stains, but they usually do not change the natural colour of teeth much⁸.

The second way is optical. Some pastes carry a blue pigment that settles on the tooth as a thin film, so that yellow reads as less yellow; on an ingredient list it appears as CI 74160, which the European Commission's CosIng database records as a copper phthalocyanine colourant and a 2026 review identifies as blue covarine⁹¹⁰. In a laboratory study on tea-stained cattle teeth, whose funding is not stated, the blue covarine paste and a microbead paste were the only ones to change shade after a single brushing cycle, and an ordinary paste changed nothing¹¹.

Whether people can see it is disputed. A 2026 systematic review and meta-analysis of human studies reported better measured and perceived colour with blue covarine, and notes that one of the five human studies it describes was supported by a toothpaste manufacturer¹⁰. A double-blind randomised trial of 75 adults, whose funding is not stated, found no difference in shade or colour measurements between a blue covarine paste and a conventional paste at any visit, including straight after the first brushing¹². Both results stand, and they do not agree.

The third way is peroxide, and in a UK shop that can only mean a trace. The fourth is hydroxyapatite, which some brands sell as a whitening ingredient; what the studies say about it is in the questions at the end of this page.

The tubes below come from the recorded rows of this site's category review, plus one whitening paste read for this page, based on what each brand states about its own formula. This page compares ingredients and stated actions only, not clinical performance. Prices and formulations may change, always check the pack.

Two rows in that table carry the same ingredient list, word for word: Sensodyne Repair & Protect Whitening and the plain Repair & Protect, as the same retailer lists them. An ingredient list shows neither particle size nor quantities, so matching lists are not proof of matching pastes. Among the sensitivity toothpastes in the scan, hydrogen peroxide appears in one list, Colgate Max White Clinical Sensitive Protect's, and neither that listing nor the page for the Spotlight Oral Care whitening paste gives a level.

Can a toothpaste bleach a tooth in the UK?

Not a toothpaste that can legally be sold to you, and the high-peroxide pastes tested abroad did not manage it well either. British cosmetics law puts toothpaste in the oral-product category, caps that category at 0.1% hydrogen peroxide present or released, and reserves whitening products between 0.1% and 6% for sale to dental practitioners¹. The page on whether tooth whitening is legal in the UK covers who is allowed to do it.

Researchers in Brazil have tried a paste far above that ceiling. In a randomised trial of 90 participants, brushing three times a day for four weeks with a 3% hydrogen peroxide toothpaste, thirty times the UK ceiling, produced significantly less colour change than two weeks of a 10% carbamide peroxide gel at home, and both routes came with tooth sensitivity and gum irritation⁴¹. In the same randomised trial, sensitivity was less likely and less intense with the paste than with the gel, and the authors concluded that a 3% hydrogen peroxide toothpaste did not bleach effectively and should not be recommended as a bleaching treatment⁴.

An older randomised trial of 43 adults, by authors affiliated with Procter & Gamble, found that a paste with 1% hydrogen peroxide and a metal activator, ten times the UK ceiling, gave no significant colour improvement over 14 days, while a 5% carbamide peroxide tray did¹³.

Two laboratory studies from the Brazilian group explain part of the gap. In the laboratory, brushing extracted premolars with pastes declared at 3%, 4% and 5% hydrogen peroxide changed their colour less than a home bleaching gel, and the paste with an acidic pH, the 4% one, let the most peroxide through to the pulp chamber¹⁴¹⁵. In the second of those laboratory studies, the pastes also let more peroxide into the pulp chamber than 10% carbamide peroxide worn at home¹⁵. Neither study measured pain, and the authors' suggestion that peroxide in the pulp chamber could matter for sensitivity is one their laboratory model could not test¹⁴¹⁵.

Cost points the same way. A cost-effectiveness analysis that modelled costs for patients in Brazil and the United States found a 10% carbamide peroxide home gel better value than blue covarine or hydrogen peroxide toothpastes in both markets¹⁶.

A paste like that has already been stopped at the UK border. In February 2025 the Office for Product Safety and Standards published a report on an imported whitening toothpaste that claimed to contain 3% hydrogen peroxide, above the 0.1% maximum for cosmetic products, and the import was rejected¹⁷. The Journal's page on sensitive teeth after whitening is about bleaching courses; this page is about the tube.

Do whitening toothpastes make sensitive teeth worse?

Nobody has shown that they do, and very few researchers have looked. A PubMed search run for this page on the tenth of September 2026, pairing whitening toothpaste and stain removal terms with dentine hypersensitivity or tooth sensitivity and with terms for randomised or clinical designs, returned fourteen records, and none once Schiff, air-blast or tactile testing was made a required term. A broader version returned thirty-eight records, and a third search, aimed at sensitivity trials of pastes sold with a whitening or stain-removal claim, returned eight.

Among all those records, one randomised trial recruited people because they had dentine hypersensitivity and then tracked that sensitivity, with a probe and an air jet, while they brushed with a paste sold with a whitening benefit. In that eight-week double-blind trial of 121 adults in China, whose funding is not stated in its record, an 8% arginine and calcium carbonate paste described as having gentle whitening benefits reduced sensitivity to both stimuli more than a fluoride paste without arginine, and no differently from the existing arginine paste without the claim¹⁸.

Six more randomised trials and one smaller clinical study, none of them described as recruiting for sensitive teeth, recorded tooth sensitivity along the way. In a double-blind randomised trial of 60 participants, whose funding is not stated, weekly scores showed no significant sensitivity over four weeks with two whitening pastes or with a conventional one¹⁹. The blue covarine trial of 75 adults, whose funding is not stated either, recorded sensitivity and gum irritation with the pigment paste at the level of the conventional paste, and lower than with the bleaching gel¹². In a 12-week randomised trial of 46 adults whose funding is not stated, minor tooth sensitivity and oral irritation were the commonest findings, but the abstract pools them across two whitening pastes and a 6% peroxide strip, and the record lists an erratum this page has not read²⁰. In a randomised study of 56 people using a whitening toothpaste, a whitening pen or both for a fortnight, 27.8% reported tooth sensitivity during pen use, and how the group on toothpaste by itself fared on that score is not stated²¹.

The peroxide pastes are a different case. Sensitivity and gum irritation turned up with the 3% paste and, as the commonest safety findings, with the 1% paste in the manufacturer-authored trial described above⁴¹³. In a small 1998 clinical study of 30 patients whose teeth had already been bleached, and whose funding is not stated, a 3% hydrogen peroxide paste had a lower sensitivity rating than a 10% carbamide peroxide paste²². After a dentist's bleaching, sensitivity in a single-blind randomised trial of 120 adults peaked in the first 48 hours and was close to zero by day seven whichever of six pastes they brushed with, five of them whitening pastes²³.

That leaves one trial in people with sensitive teeth, whose funding is not stated, and in it the paste sold with a whitening benefit did no worse than its plain sibling¹⁸.

In the laboratory, most of the evidence runs against the folklore, with one dissent. Laboratory brushing tests on 120 pieces of human root dentine found the whitening editions of two desensitising pastes no more wearing than the ordinary editions, in work whose funding is not stated²⁴. In laboratory tests on enamel, nine desensitising and whitening pastes caused no more erosive wear than artificial saliva, although one whitening paste left the surface rougher². On cattle enamel in another laboratory study, whitening pastes wore the surface as much as, or less than, the same makers' regular pastes³. The dissent is a 2022 meta-analysis of seven laboratory studies, four of them pooled, which found that whitening toothpastes made enamel rougher and lowered its microhardness⁷. None of these laboratory results is a measurement of anybody's pain, and this page does not treat one as if it were.

Contact time is the laboratory finding worth carrying out of the shop. In a laboratory study on cattle dentine, a paste containing hydrogen peroxide and citric acid at pH 5.0 removed less dentine than a standard abrasive slurry during one brushing cycle, but more when it simply sat on the surface for an hour²⁵.

The factors that observational and laboratory research does link with sensitive teeth, acid wear, gum recession and brushing force, are set out on the pages on sensitivity that is getting worse and gum recession and brushing, and for acid the Journal's page on tooth enamel erosion.

What is RDA, and why can you not find it on the pack?

RDA stands for relative dentine abrasion: a laboratory score for how much dentine a paste wears away under a brushing machine, set against a reference slurry scored at 100²⁷. The original method measures radioactive dentine released by brushing and newer work uses a profilometer to measure the depth worn, according to a 2013 review of abrasivity testing whose funding is not stated²⁸. That review, whose funding is not stated, calls RDA a sound way to rank pastes against one another and says it is inappropriate to use RDA values alone to judge clinical safety, because wear depends on many things, including how differently people brush²⁸.

The numbers spread widely. In one laboratory comparison on cattle dentine, whose funding is not stated, scores for three regular and two whitening pastes ran from 26 to 166, and the peroxide whitening paste was not among the high ones²⁶. In another laboratory study, five hydrogen peroxide whitening pastes scored between 19 and 46 and two silica whitening pastes 80 and 111, against 100 for the reference, and the share of solid particles in each paste was what moved the score²⁷. A low score is not a safety certificate, and a high one is not a diagnosis.

It is also almost never on a UK pack. For this page every column of all 77 rows in the category review's data file was read on the tenth of September 2026, and one row publishes an abrasivity value: OraNurse's page for its Unflavoured Toothpaste gives an RDA of 29.68, for a paste positioned for people with taste or sensory sensitivity rather than sensitive teeth. The count comes from the category review of UK sensitivity toothpastes, where every row links the page it was read from. No product in the first table publishes one, including the whitening paste added for this page.

The advertising regulator sets its own bar for the word. CAP's advice on teeth whitening, dated 16 September 2025, says the ASA may expect a toothpaste claimed to be whitening to be shown to whiten perceptibly more than the same brand's non-whitening pastes, and warns that even a whitening claim in a product name could otherwise breach the advertising code²⁹. That advice rests on a 2018 ruling about a website ad for a whitening toothpaste sold for sensitive teeth³⁰. The advertiser supplied five randomised, observer-blinded stain studies; one found significantly more stain removal than the comparison pastes, four found no significant difference, none tested the paste against the brand's own non-whitening pastes, and the ASA ruled the ad misleading³⁰. In its response to the regulator, the advertiser put the paste's relative dentine abrasion score at 12, while the web page itself gave shoppers no score at all³⁰.

Where does S3 sit if you are choosing between a sensitivity paste and a stain-removing one?

On the sensitivity side of that choice, and without a published abrasivity figure. S3's declared actives are 5% potassium nitrate and two hydroxyapatites, 10% nano-hydroxyapatite and 5% biomimetic hydroxyapatite, which are inclusion levels as supplied; the active hydroxyapatite content is lower, and S3 states both. Of those three, the potassium is aimed at the nerve, the nano-hydroxyapatite at the inside of the open tubule and the biomimetic hydroxyapatite at the enamel surface, and none of them covers another's job. It carries 1450 ppm fluoride as sodium monofluorophosphate, the form S3 chose because it will not react with the calcium of the hydroxyapatite sharing the tube. How the actives in sensitivity pastes do their work is covered in the Journal's piece on how sensitive toothpastes work.

S3 publishes no abrasivity figure, and in the September 2026 scan the one row that does is not a sensitivity toothpaste, so this page cannot rank S3 against any other paste on the shelf for wear, and it does not. In the one laboratory comparison of six desensitising pastes found for this page, whose funding is not stated, the paste containing zinc-carbonate hydroxyapatite wore away the most dentine³¹. S3 contains hydroxyapatite too, and nobody has published an abrasivity measurement of S3. The page on hydroxyapatite every day and abrasivity looks at that ingredient in more depth.

Some hydroxyapatite pastes sold with a whitening claim, such as the US-listed Davids paste in the first table, carry no fluoride. S3 does not ask anyone to give up decay protection, and that is where it parts company with those fluoride-free hydroxyapatite pastes.

For a tooth that already twinges, what matters is which job the tube is being bought for. A sensitivity toothpaste is judged on sensitivity, over weeks of use. A real change of shade is a different job, and the NHS and the Oral Health Foundation both send anyone who wants one to a registered dental professional³²⁸. The hub page on how to whiten sensitive teeth walks through that route.

What should someone with sensitive teeth avoid?

Five things, and most of them concern peroxide, acid and force, not the word on the tube.

Peroxide toothpastes bought online or from abroad. A paste above 0.1% hydrogen peroxide may not be sold to the public here, one has already been turned back at the border, and the trials above found that stronger pastes did not bleach well and still brought sensitivity and gum irritation¹¹⁷⁴. The NHS warns that pharmacy and online whitening kits may harm teeth and gums³². Spotlight Oral Care states "Zero sensitivity" among the highlights of its UK page for a whitening toothpaste whose list includes hydrogen peroxide without a level; this page reports the statement and has not tested it.

Acidic formulas and long contact, on laboratory evidence only. The acidic 4% peroxide paste let the most peroxide into the pulp chamber of extracted teeth, and a peroxide paste with citric acid took more dentine when it sat on the surface for an hour than during a brushing cycle¹⁵²⁵. Brush with a whitening paste for the normal time and spit it out; there is no reason to hold it against the teeth.

Fluoride-free stain powders and pastes. Several products sold for whitening, charcoal powders among them, carry no fluoride, and the page on charcoal toothpaste and sensitive teeth sets out what that trade costs.

Pressing harder to shift stain. Brushing force changes what happens at an exposed dentine surface in laboratory models, and the page on gum recession and brushing covers soft bristles and technique.

Using a stain-removing paste to cover up a tooth that has changed. Sensitivity confined to one tooth, pain that started suddenly, pain that is severe, or twinges still there after a few weeks are, for the Oral Health Foundation, reasons to see a dentist, because decay, a crack, gum problems or infection can sit behind them³³. A whitening paste changes none of those.

Frequently asked questions

Does S3 publish an RDA figure?

No. S3 publishes its full ingredient list and its declared active levels but no relative dentine abrasion figure, so neither this page nor anyone holding the tube can put S3 on the scale. If abrasivity matters to you, ask any brand, S3 included, for the figure and the method behind it, meaning the radiotracer test or profilometry, the reference used and the number of brushing strokes, because a 2013 review whose funding is not stated warns that the score alone cannot settle clinical safety²⁸.

Does blue covarine whiten teeth or only look like it?

It changes how the tooth looks rather than what it is made of: the pigment, listed as CI 74160 and recorded in CosIng as a copper phthalocyanine colourant, sits on the surface as a thin blue film¹⁰⁹. A 2026 meta-analysis of human studies reported better measured and perceived colour with it, with one included study supported by a toothpaste maker, while a double-blind randomised trial of 75 adults, whose funding is not stated, found no difference from a plain paste at any visit¹⁰¹².

Is a "sensitive + whitening" toothpaste a compromise?

Not on the evidence so far, which is thin. In a laboratory study whose funding is not stated, the whitening versions of two desensitising pastes wore root dentine no more than the plain versions²⁴. In the one randomised trial of a whitening-positioned sensitivity paste in people with dentine hypersensitivity, whose funding is not stated either, it matched the existing arginine formula at every visit¹⁸.

Can I use a whitening toothpaste during a dentist's bleaching course?

Ask the dentist running the course, who knows the gel and the timetable, and expect some sensitivity whatever you brush with: the Oral Health Foundation notes that tooth whitening usually causes temporary sensitivity³³. In a single-blind randomised trial of 120 adults after in-office bleaching, the shade measured over three months did not differ between five whitening pastes and a conventional one²³. Using a sensitivity toothpaste around a course has its own page, on whitening with a sensitivity toothpaste.

Is hydroxyapatite a whitening ingredient?

It is studied as one, and the results so far are modest. A 2023 systematic review whose authors include scientists employed by Dr Wolff, a maker of hydroxyapatite toothpastes, found that hydroxyapatite products whitened enamel in vitro and in six preliminary clinical trials, and rated the clinical evidence modest³⁴. In a laboratory study on extracted teeth, calcium phosphate suspensions changed colour by small amounts, and tricalcium phosphate did more than hydroxyapatite³⁵. How hydroxyapatite is being tested during bleaching is on the page on nano-hydroxyapatite and whitening sensitivity.