How long does nano-hydroxyapatite toothpaste take to work on sensitivity?

Nano-hydroxyapatite toothpaste takes about two weeks to produce a difference a trial can measure: two weeks is where the two cleanest trials, both double-blind and both with a placebo arm, took their first reading and found one¹². That is a schedule rather than a discovery. Two weeks is when researchers first looked, not when the paste first did anything, and the days before that first look have almost never been measured at all⁷. After it the calendar thins quickly: four weeks is where most of the evidence stops, and eight weeks is the edge of it⁸. S3 Sensitivity Science™ is judged at two to four weeks, the window the published trials of its actives use. Three trials did read earlier than a fortnight, one of them immediately after a single brushing, and each of the three carries a caveat worth knowing before you lean on it¹³¹⁴¹⁸. The rest of this page is a list of which day was read, in which trial, with which instrument — and of the days nobody has read.

What was checked22 peer-reviewed studies, S3's own consumer trial (ADSL, 2026), and guidance from the Oral Health Foundation and the NHS

When did the trials actually take their readings?

Mostly at two weeks and four, sometimes at eight, rarely later. The table below is every hydroxyapatite sensitivity trial in this library set out by the days on which somebody took a reading, because that column decides what any of them can tell you. The last column is the one that matters most: it is where the evidence stops.

The final row is deliberately fenced off. It is a panel of people answering questions about their own teeth with nothing to compare against, which is a different kind of evidence from the twelve rows above it, and this page treats it that way.

Two things stand out. The first is how short the last column is: ten of these twelve trials had stopped reading by week eight, and only two carried on past it. The second is how often the answer in the sixth column is "none": in four of these trials the hydroxyapatite arm never separated from what it was tested against. The eight-week trial most often quoted for the ingredient, which was sponsored by the maker of the pastes it tested, had no placebo arm and no plain-fluoride arm at all, so every one of its four toothpastes falling from baseline is not by itself evidence that any of them did the falling³.

A smaller and more recent randomised trial makes the same point from the other end. Thirty-nine adults, no blinding, a hydroxyapatite paste against a potassium nitrate paste, readings at two weeks and four: both fell significantly from baseline, neither pulled ahead, and the authors say plainly that a trial that size can neither establish nor exclude a difference between the two actives¹⁵. Meanwhile a six-arm randomised trial of 164 adults with no hydroxyapatite in it at all shows why "compared with what" is the question underneath "how long": its plain fluoride control improved too, but only by week eight, and only on two of its three measures¹⁶. Whatever a sensitivity toothpaste is doing at week four, some of it is what any toothpaste does.

What do the pooled reviews say, and which window do they cover?

Each of them covers a different slice of the calendar, which is why they disagree without contradicting each other.

The 2019 systematic review of 74 randomised trials is the one built around time, and its finding is blunt: the only treatments with a significant reduction inside the first seven days are the ones a dentist applies in the chair, and hydroxyapatite first appears in the up-to-one-month group⁷. The 2026 network meta-analysis of 93 randomised trials is the largest, and its networks all sit at a single visit, two weeks; it does not reach past that point, and says so⁶.

Does it work faster on some triggers than others?

On the published evidence it works on some triggers and not, so far, on the one most readers care about. The 2019 meta-analysis of six four-week randomised trials separated the stimuli, and nano-hydroxyapatite came out ahead of its comparators on the evaporative test (a blast of air) and on the tactile test (a probe), while on cold there was no difference at all⁴. That is worth sitting with. In the six trials that meta-analysis pooled, the outcome measured most often is a puff of air from a dental syringe; the outcome that sends people to a chemist is a mouthful of iced water⁴.

Two trials do report a cold result: the double-blind trial of 105 adults found lower cold-air sensitivity at two weeks and at four¹, and its companion trial with a 10-minute gel found the same². Both used a fluoride paste and a placebo as comparators, which makes them the cleanest cold readings in the set.

The ranking also moves with the week. In the network meta-analysis of 30 randomised trials, nano-hydroxyapatite held the highest probability of being the most effective toothpaste at two weeks and at four, and arginine took that position at eight⁵. A ranking probability is not a demonstrated difference — the same analysis found no significant pairwise superiority between the actives, and it measured a large placebo effect in this kind of trial⁵. Sealing and calming run on different clocks in the first place, and neither active does the other's job. The page on how nano-hydroxyapatite seals a tubule follows the mineral side of it in detail.

The mineral itself has been measured on a shorter timescale than the symptom, in a different kind of study¹⁹. In an in-situ study, 80 people wore appliances carrying blocks of human dentine while brushing twice a day; by day seven the nano-hydroxyapatite dentifrices had laid down significantly more precipitate and left significantly fewer open tubules than a monofluorophosphate one, and the amount at day fourteen was not significantly different from the amount at day seven¹⁹. Mineral on a block of dentine is not comfort in a tooth, and this page does not treat it as such.

What happens after eight weeks, and after three months?

Very little is known, and what is known is not flattering. The longest randomised trial of hydroxyapatite pastes for sensitivity in this library ran three months, in 28 adults, and it began with three professional applications in a surgery rather than with a tube from a shop; at three months none of its regimens differed from the others¹¹. The only trial that reads at twelve weeks is manufacturer-funded, and it spends four of those weeks with the participants off the product¹⁰. Beyond that there is nothing.

The one study in the library that did keep reading to eight weeks against a control group is a warning rather than a comfort, and it is not a toothpaste¹². Hydroxyapatite applied in the surgery to 486 teeth in 45 patients beat distilled water and no treatment at every reading from day one to week four in a randomised trial, and by week eight the four groups no longer differed at all¹². In children with molar-incisor hypomineralisation, a double-blind non-inferiority trial ran 56 days and could not show that a microcrystalline hydroxyapatite toothpaste matched an amine fluoride one; the trend favoured hydroxyapatite, and the trial lost too many participants to prove anything either way¹⁷.

There is one long-term synthesis, and it cuts the other way. A 2025 systematic review and meta-analysis pooled only randomised trials with at least six months of follow-up — 22 of them, 15 with usable data — and reported that while several desensitising agents held a significant benefit that far out, adhesive systems and calcium-phosphate-based agents did not reach statistical significance⁹. Hydroxyapatite belongs to that class. The abstract does not name it separately and the full text is not open access, so this page cannot tell you whether a hydroxyapatite trial was in that pool at all⁹. What it can tell you is that the strongest long-term evidence the field has does not currently support the class.

Two independent readings of the wider literature say the same thing in their own words. A 2026 systematic review of 31 randomised trials of the newer desensitising approaches notes in its discussion that nano-hydroxyapatite and bioglass formulations are widely marketed as superior options while long-term clinical data on them remain sparse⁸. A 2025 narrative review of 15 recent clinical studies of hydroxyapatite oral care products closes by asking for standardised long-term studies before the clinical relevance of the ingredient can be treated as settled²². Neither review has a product in this argument.

What happens if you stop?

One manufacturer-funded trial has now measured it, and the answer is split by instrument¹⁰. In a twelve-week double-blind randomised trial of 129 adults, funded by the toothpaste's maker and with four of its six authors employed by that company, participants used a hydroxyapatite toothpaste for eight weeks and then switched to a plain fluoride toothpaste for four¹⁰. At the twelve-week visit the advantage on the air-blast score over the placebo group was still there. The tactile threshold was not: in the same manufacturer-funded randomised trial the improvement over placebo fell from 100.85% at eight weeks to 14.56% at twelve, and the authors put that down to the surface mineral layer not being replenished¹⁰. One more detail from that industry-funded randomised trial belongs here: neither active arm had separated from its own baseline at the first visit, two weeks in, and significance arrived only at six¹⁰.

Two cautions attach to that. The toothpaste in question carried potassium citrate as well as hydroxyapatite, so this manufacturer-funded randomised trial cannot say which of the two the surviving effect belongs to¹⁰. And it is one trial, paid for by the company that sells the product.

The comparison worth drawing is with the other actives, where the same question has been asked more carefully. In a double-blind randomised trial of 120 adults, run and funded by a toothpaste manufacturer whose own product was one of the arms, everybody went back onto a plain fluoride paste for three weeks at the end of the eight-week treatment period²⁰. In that manufacturer-funded trial all three test pastes still held a significant benefit over the control when they were read again three weeks later, keeping between roughly two-thirds and five-sixths of what they had held at the end of treatment, depending on the instrument²⁰. So the effect of a desensitising toothpaste is not a switch that flips off the day you stop. It also is not something you keep for free: the Oral Health Foundation's position is that these toothpastes work by blocking the channels in dentine and that you have to keep using them to maintain the effect²³. On hydroxyapatite specifically, one industry-funded trial and a general piece of guidance are what exist, and they do not entirely agree.

How long should you give S3 before deciding?

Four weeks, and eight before concluding anything final. That is not a house rule; it is where the trials of the actives in the tube take their readings, and judging a paste on day five is judging it at a point no study has ever measured. The tube is not a course to be finished; it is what you brush with, and the useful question is what has changed by the fourth week.

S3's own figures sit at the early end of that window, and they are self-reported, which is the thing to know about them before the number: they come from an independent third-party consumer trial of 51 adults with sensitive teeth over eight weeks. In it, 84% said their teeth felt less sensitive in trigger situations after one week. That is a panel answering a questionnaire rather than a probe reading a tooth, and the two can part company: a 2017 systematic review and meta-analysis of seven randomised, double-blind, placebo-controlled trials of desensitising mouthwashes found patients reporting significantly less everyday sensitivity at eight weeks while the tactile and air-blast measurements showed no significant difference at all²¹.

Concentration is a separate question from time. In the manufacturer-sponsored eight-week trial that put 10% and 15% nano-hydroxyapatite in the same base, and in the in-situ study that put them on dentine blocks in 80 mouths, no comparison showed the higher percentage buying a quicker result³¹⁹. What percentage does buy is set out on the page on what a toothpaste needs. Whether the ingredient works at all, weighed across the whole evidence base instead of by timepoint, is a page of its own. The same clock question for the potassium side of the formula is answered on its own timeline page, and the case for why relief that builds over weeks is a feature, not a flaw is argued elsewhere. The Journal's explainer on how sensitive toothpastes work and why yours might not be working for you covers the same ground for a first-time reader.

What if nothing has changed after eight weeks?

Then stop giving the toothpaste more time and give a dentist a look, because the list of things that mimic dentine sensitivity is long and none of them is a toothpaste's job. The Oral Health Foundation's own reasons to book an appointment are short and specific: pain that is severe, sensitivity that has gone on more than a few weeks, a single tooth rather than several, or a twinge that arrived suddenly²³. Any of those can turn out to be decay, a crack, gum disease or an infection. The NHS puts the threshold for toothache at two days, and adds fever, pain on biting, red gums or a bad taste as reasons not to wait²⁴.

There is also a duller possibility. A sensitivity toothpaste addresses an over-ready nerve and open dentine tubules, and if neither is what is wrong with your tooth then no amount of patience will help. Guidance on switching, and on what a fair test looks like is set out separately, as is what daily care can realistically do over years.

Frequently asked questions

How quickly does S3 work?

Judge it at four weeks and give it eight before deciding, because that is where the published trials of its actives take their readings. In S3's own consumer trial of 51 adults over eight weeks, run by an independent agency, 88% said their teeth feel less sensitive at four weeks. No toothpaste in this evidence base has been shown to work on the first day: the earliest reading in most randomised trials is at two weeks, and the handful of earlier readings come with conditions attached¹²¹³.

Can a toothpaste work on the first day?

One randomised trial of 100 adults did measure a significant fall in air-blast and VAS scores 30 seconds after a single application of a biomimetic hydroxyapatite toothpaste or mouthwash, straight after a professional clean¹⁸. It is a real result and it is narrow: both arms were hydroxyapatite products from the same maker, so there was nothing to compare against, the reading was taken once and never repeated, and nothing in the trial says the effect was still there the following morning¹⁸. A separate examiner-blind trial found a reduction immediately after brushing with a toothpaste combining potassium nitrate, aluminium lactate and hydroxyapatite, which cannot tell you which of the three did it¹³. Treat both as evidence that something happens at the surface quickly, not as a promise about your week.

Does it stop working if you stop using it?

Partly, on the one trial that has asked. Four weeks after switching off a hydroxyapatite-and-potassium-citrate toothpaste, the air-blast benefit in a 129-adult randomised trial funded by the manufacturer was still measurable while the tactile benefit had largely gone¹⁰. The Oral Health Foundation's general advice is that sensitive toothpastes need continued use to maintain their effect²³, and for the other desensitising actives a manufacturer-funded randomised trial of 120 adults found most of the benefit surviving three weeks off the product²⁰.

Should I give it longer than eight weeks?

You can, but you would be past the evidence. One hydroxyapatite trial in this library reads at three months, and one manufacturer-funded trial reads at twelve weeks¹⁰¹¹. The systematic review and meta-analysis restricted to trials of six months or longer did not find a significant long-term benefit for the class hydroxyapatite belongs to⁹. If eight weeks of consistent twice-daily brushing has changed nothing, the useful next step is a dentist rather than a ninth week.

Does a higher concentration work sooner?

Nobody has shown that it does. The eight-week randomised trial sponsored by the maker of the pastes it tested compared 10% and 15% nano-hydroxyapatite in the same base and found them apart on one of eight comparisons, at week six on the cold stimulus only³. The in-situ study of 80 people found no significant difference between the two concentrations on any occlusion measure at either reading¹⁹. What percentage buys, and what it does not, has its own page.