Two problems in one tooth: why sensitivity needs the nerve and the tubules addressed together.

A sensitive tooth generally has two faults running at once — dentine tubules that are open to the mouth, and a nerve that has become easy to set off — and the ingredients that deal with them are different ingredients doing different jobs. "Repair" in the sense most people mean when they type it, new dentine grown back where it went, is not something any toothpaste does; what a mineral paste does is put hydroxyapatite into the open channels and onto the surface, which can be measured, while a potassium salt works at the far end of the same channel on the nerve. S3 Sensitivity Science™ is one of the pastes that carries both jobs in one tube — 5% potassium nitrate, nano-hydroxyapatite and biomimetic hydroxyapatite at 10% and 5% inclusion levels as supplied, and 1450 ppm fluoride — all of it readable on the back of the pack.

What was checked24 peer-reviewed studies, product information as published by each brand, the Oral Health Foundation and the NHS

What does "just numbing the pain" actually mean?

It means working on the nerve end of the problem and leaving the opening alone. Dentine is threaded with microscopic tubules that run from the outer surface towards the pulp; when enamel wears through or the gum recedes, those tubules are open to whatever is in the mouth, and cold, sweet or a puff of air moves fluid inside them. A potassium salt does nothing to the opening. It is thought to raise the potassium concentration around the nerve endings until they stop firing so readily, which changes the alarm rather than what keeps setting it off.

The ingredient people mean by numbing is potassium nitrate, in desensitising toothpastes since the 1970s and in placebo-controlled trials since the 1990s. The evidence sits in an unusual shape: a 2000 review identified 16 double-blind randomised trials of potassium toothpastes and reported that every one of them found significant reductions in tactile, air-blast and subjective sensitivity, and the same review states that the proposed mechanism has never been confirmed in intact human teeth¹. Something reliable is happening; nobody has watched it happen. A 2025 review of the ion channels involved argues that the current working model keeps the fluid movement and adds ion-channel activity in odontoblasts and sensory neurons, and concludes that occluding tubules or desensitising dentine remains the safest practical approach². The hydrodynamic theory page takes that mechanism apart properly; the point here is only that the chain has a nerve at one end and an opening at the other, and an ingredient can be aimed at either.

Calming the nerve is a real job, honestly done. It is also, by construction, half of the tooth, and what a nerve-only toothpaste leaves undone argues that half in detail. The Journal's account of how sensitive toothpastes work, and why yours might not be working for you is the shorter version.

What could "repair" mean, and which meanings are real?

Four different things, and only two of them survive contact with the evidence. The word arrives on the reader's screen carrying the strongest of the four meanings and usually leaves a box carrying the weakest, so it is worth pulling apart before anything else on this page makes sense.

Growing new dentine. No. The clearest test of it is a 2022 comparison, three of whose authors are employed by the company developing the peptide it favours, that ran laboratory work on human dentine disks alongside an animal hypersensitivity model: after six weeks of brushing, the group given an experimental peptide toothpaste formed new tertiary dentine, and no new hard tissue appeared in the control group or in either of the two marketed desensitising toothpaste groups⁶. That declared interest is a reason to read the peptide arm carefully and no reason at all to doubt the empty control columns. Tertiary dentine is laid down by the tooth's own cells under the pulp when they are provoked; it is not a product a tube can contain.

Putting mineral into the opening and onto the surface. Yes, and it is measurable. In an in vitro electron-microscope study of 40 dentine discs, a nano-hydroxyapatite toothpaste closed about 98% of the tubules after seven days of application⁷. Surface deposition has been seen in people rather than only on discs: in an eight-week comparative in vivo study, teeth later extracted from users of a zinc-carbonate hydroxyapatite toothpaste carried a hydroxyapatite-rich coating, while teeth from users of a potassium nitrate and fluoride toothpaste showed no appreciable change to the enamel surface⁸. Five subjects per group and no randomisation makes that a demonstration of the deposit rather than a measurement of it⁸. The ranked review of the microscope evidence goes ingredient by ingredient.

Remineralising softened enamel. Yes, and it is a separate literature with its own reviews — in vitro, nano-hydroxyapatite toothpastes remineralised bovine dentine lesions more than an amine fluoride toothpaste⁹. This page does not re-argue it; the pages on remineralising softened enamel and on whether enamel can be rebuilt do.

Putting the tooth back to how it was before the gum receded or the enamel wore. No. Recession is lost gum position and wear is lost tissue, and a toothpaste changes neither.

Why is a sensitive tooth two problems and not one?

Because one paste can quieten the pain and leave the opening exactly as it found it, and a single study caught both halves of that at once³. The trial ran four weeks, double-blind and randomised, in 30 adults: 5% potassium nitrate beat a non-desensitising toothpaste on the air-blast and cold-water measures, and under the microscope afterwards its tubules were as open as they had started³. Relief without sealing. That is the two-problem argument in a single paper, and it is the reason this page exists.

The tidy version of that story is not quite safe, and the honest treatment is to print the contradiction. A 2017 in vitro electron-microscope study of 90 dentine samples reported that a potassium nitrate toothpaste applied twice daily did produce detectable tubule occlusion at three, seven and 14 days, and its authors say randomised trials would be needed to know whether that means anything in a mouth⁴. Two laboratories, two answers, no resolution: whether potassium nitrate occludes at all is unsettled, and the page on whether potassium nitrate blocks tubules is where that argument is kept.

There is also a direct human comparison of the two approaches, and it is small. In a four-week randomised trial of 39 adults, a hydroxyapatite toothpaste and a potassium nitrate toothpaste both relieved air-blast sensitivity significantly at two and four weeks and were comparable to each other⁵. With 39 participants, no blinding and no negative control, that trial can neither establish nor rule out a difference between the two actives⁵. It is worth knowing that the two actives have been put side by side in people. It is not worth more than that.

How much evidence is there for treating both at once?

Less than the shelf implies, and it is worth counting rather than asserting. The strongest single result is one node in a 2020 systematic review and network meta-analysis of 125 randomised trials and 12,541 patients: against a fluoride toothpaste, potassium combined with hydroxyapatite showed large beneficial effects on both the tactile stimulus (standardised mean difference 2.47) and the air stimulus (2.44), with moderate to high certainty¹⁰. In that same network meta-analysis, potassium on its own appears for the tactile stimulus only, at 1.23, and calcium sodium phosphosilicate ranked most beneficial across all three stimuli¹⁰.

Then the more recent and more conservative synthesis. A 2026 systematic review and network meta-analysis of 93 randomised trials and 9,548 participants has no potassium-plus-hydroxyapatite node at all, and where its nodes exist they are thin: nano-hydroxyapatite rests on two studies, potassium with or without fluoride on one, potassium with stannous fluoride on one, against ten studies for stannous fluoride¹¹. Its authors conclude that stannous fluoride and arginine toothpastes should be first-line self-care, with the choice between actives guided by patient preference, tolerability and availability rather than an expectation of superior efficacy¹¹. Neither of those two classes is in S3. That sentence costs this page something and it stays, because a reader deciding what to buy deserves the same review the authors wrote rather than the half of it that suits a brand. The same review records that most of the trials it pooled were industry funded, including 96% of the stannous studies and 76% of the potassium studies¹¹. A 2023 systematic review and network meta-analysis from the same Bristol group, whose authors are affiliated with dentifrice manufacturers through advisory fees and research grants, reached a similar list of recommendable actives¹².

The reviews also disagree with each other, which is the state of the field rather than a flaw in the reporting. A 2019 network meta-analysis of 30 randomised trials gave nano-hydroxyapatite the highest probability of being most effective at two and four weeks, and in the same network found potassium toothpaste not significantly different from placebo¹³. A 2018 systematic review and meta-analysis of 53 randomised trials in 4,796 patients places potassium above a toothpaste with no active at all¹⁴. A 2015 meta-analysis of 31 randomised trials put potassium toothpastes at a standardised mean difference of −1.28 against placebo, with heterogeneity between trials so high that the pooled figure should be read as a direction rather than a size¹⁵. For the mineral side, a 2023 systematic review and meta-analysis of 44 clinical trials, whose authors include scientists employed by a German maker of hydroxyapatite toothpastes, found hydroxyapatite products reduced sensitivity by 39.5% against placebo and by 23% more than fluoride toothpastes, and found no significant difference against other desensitising agents¹⁶.

Has a toothpaste with both actives been tested against the same paste with each alone?

Half of it has, once, and the half that matters most has not. That is the question underneath the reader's question, and the answer is worth more than any claim on this page, so here is exactly what was searched. Two PubMed searches were made for this page in September 2026: one for factorial, two-by-two, component, single-agent, additive or synergistic designs in toothpaste work on dentine hypersensitivity, and one pairing potassium nitrate with hydroxyapatite and the word "alone". The second returned nothing at all. The first returned ten records, none of them a factorial design of the two actives, and one of them changed how the absence should be described.

The nearest thing to the missing study is one arm inside an eight-week double-blind randomised trial of 85 adults, funded by the company whose products it tested: a 10% nano-hydroxyapatite toothpaste ran against the same formula carrying 5% potassium nitrate as well, and the version with potassium eased cold-stimulus sensitivity more at weeks two, four and six and air-stimulus sensitivity more at weeks four, six and eight¹⁷. That is one half of the design — the pairing against the mineral alone. Nobody has run the other half, the pairing against the potassium alone, and that trial had no placebo or plain-fluoride arm to size any of it against.

Elsewhere the component design does exist, for a different second ingredient. A 2026 double-blind randomised trial of 127 adults, funded by the manufacturer whose employees wrote it, set a toothpaste with 5% potassium nitrate alone against a toothpaste with 5% potassium nitrate plus 2.18% aluminium lactate and against a placebo; both test pastes beat the placebo from two weeks onward, and the two-active paste beat the one-active paste at four weeks on air stimulation and at eight weeks on both stimuli¹⁸. So the design is possible and it has been run once — for a different second ingredient, by the company that sells it. A 2021 UK examiner-blind randomised trial pointed the same way, with a paste combining potassium nitrate, hydroxylapatite and aluminium lactate reducing sensitivity more than a potassium-nitrate control toothpaste at every time point over 14 days; aluminium lactate sits in that formula too, so the gain cannot be credited to the hydroxyapatite¹⁹.

The counterweight belongs on the page as much as the support does. A 2024 study paired laboratory work with a triple-blind randomised split-mouth clinical phase on a paste combining 8% L-arginine, calcium carbonate and potassium nitrate: in vitro it significantly reduced dentine permeability and occluded most tubules, and in the clinical phase, with 16 to 20 patients, it did not separate significantly from its placebo on cold, tactile or spontaneous sensitivity at any point out to three months²⁰. Both arms improved. Two other pieces sit either side of that. Nano-hydroxyapatite and 5% potassium nitrate have been formulated together and characterised in vitro on 24 extracted human teeth, where both concentrations tested raised surface hardness and partly restored mineral content²¹ — mechanism, on extracted teeth, and nothing more. And a 2024 double-blind randomised trial of 100 adults, run without industry money, found that a toothpaste combining hydroxyapatite, potassium citrate, zinc citrate and 1450 ppm fluoride left patients reporting less sensitivity at four weeks than a fluoride-only toothpaste, on a custom questionnaire rather than an instrument²².

Put together, the case for treating the nerve and the tubule in one paste is a mechanistic case with real but thin direct support. Each lever demonstrably leaves undone what the other one does; the pairing has one favourable node in a network meta-analysis of 125 randomised trials¹⁰ and one manufacturer-funded arm in which the version carrying potassium outscored the version without it¹⁷; and the study that would settle it — one formula, its actives removed one at a time, against a control — has not been published. This page will not tell you the combination has been shown to work better, because the literature does not yet say so. The page on single-active versus multi-active toothpastes works through what changes when mechanisms are combined, and tubule occlusion versus nerve desensitisation sets the two approaches out as a taxonomy.

Why can't every toothpaste just contain both?

Because the tube is a chemistry problem before it is a marketing decision. Potassium nitrate needs water to dissolve into and travel in, and that is the whole of it: a water base is what allows S3 to hold potassium nitrate and hydroxyapatite in the same tube, and anhydrous formulas built around bioactive glass have no such option. Fluoride raises a second constraint: S3 keeps 1450 ppm fluoride as sodium monofluorophosphate, chosen because it does not react with the calcium in hydroxyapatite inside the tube.

You can read that constraint straight off the shelf, based on what each brand states about its own formula. Sensodyne Repair & Protect declares calcium sodium phosphosilicate in a glycerin and PEG base with no water in the ingredient list, while Sensodyne Daily Care Original declares potassium nitrate in a water-based paste — one brand, two tubes, one job each. Nor is combining the two an S3 invention: Spotlight Oral Care's Sensitivity + Rebuilding Pro declares potassium citrate, hydroxyapatite and sodium fluoride together in a water-based formula. This page compares ingredients and stated actions only, not clinical performance, based on what each brand states about its own formula. Prices and formulations may change; always check the pack.

What does S3 do about each of the two problems?

It puts an active against each of them and states the level of each. Potassium calms the nerve, nano-hydroxyapatite is sized to go into the tubule, and biomimetic hydroxyapatite works on the surface, so none of the three is asked to do another's job. The two hydroxyapatites are not a duplication: the nano form is under 100 nm, rod-shaped and crystalline, and the biomimetic form is around two microns and less crystalline, which is why the two forms of hydroxyapatite have a page of their own.

The levels are the argument, and they are also its limit. S3 is formulated at inclusion levels that match the published studies, 5% potassium nitrate and 10% nano-hydroxyapatite solution; 5% is the potassium nitrate dose the placebo-controlled trials used, and the only dose for which "clinically proven" belongs anywhere near this ingredient. Those percentages are inclusion levels of the ingredient as supplied, and the active hydroxyapatite content is lower — S3 states both. Fluoride stays in the formula rather than being traded away for the mineral, which is what separates it from the fluoride-free hydroxyapatite pastes. What none of that amounts to is a trial of the finished product against anything, and this page is not going to imply one. The honest form of the claim is: each active is present at a level the literature has tested, and the case for having them in one tube is mechanistic.

What does neither job do?

They do not switch off the causes. The Journal's page on the two causes of sensitive teeth sets out why a tooth becomes sensitive in the first place: gum has receded or enamel has worn, and the receding and the wearing carry on regardless of what is in the tube — brushing too hard, acid from fizzy drinks, juice, citrus or the stomach, brushing straight after acidic food, and gums that recede with age are all on the Oral Health Foundation's own list²⁵. Grinding and gum disease sit outside a toothpaste's reach altogether.

Duration is the other gap. A 2025 systematic review and meta-analysis of 22 randomised trials with at least six months of follow-up found significant long-term pain reductions for several desensitising agents, with the largest effects for glutaraldehyde and low-level laser therapy, and reported that calcium-phosphate-based agents did not reach statistical significance²³. That review reports a class rather than naming hydroxyapatite, and it is the strongest long-term evidence anyone has — which is why it is printed here rather than left out. The page on what long-term daily care can realistically do takes that further. A 2017 UK guideline review for general dental practice puts the field's own position plainly: there does not currently appear to be one ideal desensitising agent that can be recommended²⁴.

And some pain is not this problem at all. See a dentist when the pain is severe, when sensitivity has run on for more than a few weeks, when it is confined to a single tooth, or when it arrives suddenly, since those patterns point at decay, a crack, gum trouble or infection rather than open tubules²⁵. The NHS advises a dental appointment for any toothache lasting more than two days, and A&E if swelling reaches the eye or neck or makes breathing or swallowing difficult²⁶. No toothpaste, whichever end of the channel it works on, is the right first step for those.

Frequently asked questions

Does S3 repair teeth?

Not in the sense of growing new dentine, and no page in this project will say otherwise. What S3 does that the word "repair" fairly covers is deposit hydroxyapatite in the open tubule and on the enamel surface, which is what electron microscopy shows for hydroxyapatite toothpastes generally⁷⁸, and contribute to remineralising softened enamel, which is a separate body of work with its own reviews⁹. What it does not do is put back gum that has receded or enamel that has worn away. When the word appears on a box, the second of those four meanings is usually what is meant, and the fourth is usually what the reader hears.

Has a toothpaste with both actives been compared with one containing each active alone?

Half of it has. In an eight-week double-blind randomised trial of 85 adults, funded by the company whose products it tested, a 10% nano-hydroxyapatite toothpaste carrying 5% potassium nitrate eased cold and air sensitivity more at several time points than the same formula without the potassium¹⁷. Nobody has run the mirror image, the pairing against the potassium on its own, and nobody has run either against a control. Two searches for factorial or component designs made for this page in September 2026 found no such trial; the only complete component design they returned pairs potassium nitrate with aluminium lactate instead, in a manufacturer-written trial of 127 adults¹⁸. That is a different second active, and one trial.

Can a toothpaste grow new dentine?

No. In a six-week animal model paired with laboratory work, reported by a team partly employed by the peptide's developer, new tertiary dentine appeared only in the group given an experimental peptide toothpaste; the control and both marketed desensitising toothpastes produced no new hard tissue⁶. Tertiary dentine is made by the tooth's own cells when they are provoked, so the honest description of what a mineral toothpaste does is deposition, not regrowth.

If stannous fluoride scores best in the reviews, why isn't it in S3?

Chemistry, and the reviews' own caution. The 2026 network meta-analysis rates stannous fluoride's two-week result on ten studies with high confidence and names it, with arginine, as a first-line option — while its authors say the choice between actives should follow preference, tolerability and availability rather than an expectation of superior efficacy¹¹. Building around a stannous system is a different formulation problem from building a water-based paste that carries potassium nitrate with hydroxyapatite, and nothing here claims the reverse comparison in S3's favour.

Is a toothpaste that only calms the nerve doing nothing useful?

It is doing half of a real job, and doing it on decent evidence — 16 double-blind randomised trials of potassium toothpastes all reported significant reductions in tactile, air-blast and subjective sensitivity in a 2000 review¹. S3 contains potassium nitrate for exactly that reason. The argument on this page is not against calming the nerve. It is that the opening stays open unless something is put into it.