Can teeth remineralise?
Yes, at the earliest stage — enamel that has lost mineral can take it back from saliva. But once a cavity has broken the surface, no diet, paste or oil will close it.
Yes — at the earliest stage. Enamel that has lost mineral can take it back from saliva, and that happens in your mouth every day without you doing anything. But once a cavity has broken through the surface, no diet, paste, oil or supplement closes it. The wellness internet routinely blurs those two situations together. This page is the line between them.
The tide that runs all day
Enamel is not a sealed shell. It is a mineral surface sitting in a fluid that is constantly either dissolving it or rebuilding it, and which of those is happening depends almost entirely on how acidic your mouth is at that moment.
Eat or drink something, and the bacteria in dental plaque ferment the sugars in it and produce acid within minutes. Below roughly pH 5.5 — the critical pH for enamel — the surface starts giving up calcium and phosphate. That is demineralisation.
Then saliva goes to work. It buffers the acid, washes the sugar away, and it is itself supersaturated with calcium and phosphate, so as the pH climbs back the mineral moves the other way and returns to the enamel. That is remineralisation. The oral microbiome guide covers the bacterial half of this in full.
Both directions run every day of your life. Decay is not an event; it is what happens when the balance of that cycle tips one way for long enough.
Frequency beats quantity
This is the practical consequence, and it is the thing most people have backwards.
Each eating episode costs you a recovery window — the time it takes saliva to bring the pH back up. Six biscuits eaten together cost you one window. Six biscuits eaten one an hour cost you six, and your teeth spend most of the day below the line, as in the lower half of the figure above.
A 2026 systematic review in BMC Public Health comparing the two directly concluded that “the frequency of sugar consumption was a stronger determinant of caries development than the total quantity consumed”. A 2019 review in Caries Research put the practical version more bluntly: reducing the amount without reducing the frequency does not appear to work as a prevention strategy.
It is not quite unanimous. A 2017 modelling study in Scientific Reports found that amount and frequency “may combine” to drive the process rather than frequency dominating. The weight of the clinical evidence favours frequency; the honest summary is that frequency is the bigger lever, not the only one.
Which means the sipped drink is worse than its sugar content suggests, and grazing is worse than the same food eaten at once.
What remineralisation can reverse — and what it cannot
Everything above has a hard limit, and it is the single most important thing on this page.
An early lesion — the chalky white spot you sometimes see near the gumline — has lost mineral beneath a surface layer that is still intact. That surface is a scaffold. Mineral can be redeposited into the porous body underneath it, and the lesion can genuinely arrest or reverse. This is real, it is well established, and dentists use it deliberately.
A cavity is what happens when that surface layer collapses. There is now a hole: no scaffold, and no biological mechanism that regrows enamel, because enamel is not living tissue and the cells that made it are gone once the tooth has erupted. A cavity does not heal. It gets cleaned out and filled, or it gets bigger.
The cross-section on the hydroxyapatite toothpaste page shows exactly where that boundary sits.
Where the misleading claim lives
Almost every “I healed my cavities naturally” story trades on this boundary. Early lesions really can reverse, and someone who arrested one and describes it as healing a cavity is not necessarily lying — they may simply never have had one.
That is what makes the claim so durable: there is a true statement underneath it. The problem is what a reader does with it — deciding that the hole in their own tooth can be managed with diet, and waiting, while it gets deeper and considerably more expensive.
What actually helps
- Fluoride toothpaste The best-evidenced remineralising agent there is, by a wide margin, and the least glamorous. Twice a day, and spit rather than rinse — rinsing washes away what you just applied.
- Cutting the number of eating episodes Not necessarily the amount. This is the highest-value change most people can make, and it costs nothing.
- Saliva flow Saliva is the delivery system for the whole process; without enough of it the tide simply stops coming back in. Medication is the most common cause of a dry mouth — see what causes dry mouth.
- Chewing sugar-free gum after meals Worth doing, but be clear about why: the reliable mechanism is that chewing stimulates saliva. The evidence for xylitol specifically is weaker than its marketing. The strongest Cochrane finding is low quality evidence that xylitol added to fluoride toothpaste beat fluoride toothpaste alone in children’s permanent teeth, with the rest graded low to very low.
- Hydroxyapatite toothpaste A genuine remineralising agent with real but younger evidence, most of it manufacturer-funded. The full assessment is here.
What does not do what it claims
These are claim types rather than particular sellers, and they share one move: applying a true statement about early lesions to a situation it does not cover.
- Diets and protocols that promise to reverse existing cavities. A good diet genuinely shifts the balance and prevents new damage. It does not close a hole.
- Oil pulling. It does not remineralise anything — see whether oil pulling does anything.
- Charcoal and abrasive “detox” pastes. These work in the opposite direction: they remove surface material rather than depositing it. See natural teeth whitening.
- Anything swallowed, sold as rebuilding enamel. Remineralisation happens at the tooth surface, from saliva in contact with it. A mineral going down your throat is not where the work is done.
When to see someone
Anything that looks like a hole, a white spot that has turned brown, sensitivity that has changed, or any pain at all.
The reason to go rather than experiment is precise: an early lesion and a cavity are the two cases this whole page distinguishes, one is reversible and one is not, and you cannot reliably tell which one you have by looking. A dentist can — and can remineralise and monitor an early lesion deliberately instead of filling it. That option only stays open while the surface is intact.