A Veneer That Keeps Coming Off Is Telling You Something

Keep the veneer clean and dry, and do not glue it back yourself. Household and over-the-counter adhesives are not made for the mouth and they contaminate the fitting surface, which can turn a reusable restoration into one that has to be remade. Avoid biting on the exposed tooth, and expect it to feel rough or sensitive to cold, since that surface was prepared to be covered. Then arrange an examination: the veneer is often reusable, but that is decided after looking at the restoration, the tooth and the reason it came off.

Some situations should not wait. Significant pain, swelling, a bad taste or smell around the tooth, or a tooth that has itself become mobile all point to something beyond a detached restoration.

A single debond does not necessarily mean anything was done badly. Biting onto something unexpectedly hard, a knock to the face, an old filling that has finally given way: once can have an ordinary cause.

What makes repetition different is how uncommon debonding is to begin with. In an eleven-year prospective trial of 384 feldspathic veneers in 104 patients there were 19 failures in total, and only three were debonding. Fifteen were fractures. Coming unstuck is not the usual way a veneer ends its life, and the same unit doing it twice is unlikely to be the same coincidence happening twice.

The second debond is not a repair request. It is a finding. That does not make re-bonding the wrong treatment: once the cause is established, re-bonding the same veneer is often exactly right. What changes is the order. Assessment first, adhesive second.

Ceramic is bonded to the tooth, and the tooth is not one material. Enamel is hard, mineral and largely inert, and it etches to a predictable pattern that resin locks into. Dentin is living, wet and full of microscopic tubules carrying fluid outward. Bonding to it is routine, but technique-sensitive in a way that bonding to enamel is not.

The clinical evidence points consistently the same way. A 2025 systematic review and meta-analysis reported survival and success of about 99% for veneers bonded to enamel, against roughly 91% survival and 74% success where dentin exposure was severe. A separate series of 672 veneers followed for one to fifteen years found dentin exposure carried around three and a half times the odds of failure.

Those figures need reading carefully. They are study-level rates covering every kind of failure together, and debonding is only a minority of them. They establish that the bonding substrate matters a great deal to how veneers perform. They are not the odds that your veneer will come off again, and no published figure can tell you that.

Where dentin is unavoidable there is a well-studied way of managing it. Sealing the exposed dentin immediately at the preparation appointment, rather than weeks later at cementation, gives the adhesive a mature, uncontaminated surface to bond to. In that same eleven-year trial, among teeth with more than half the surface in dentin, survival was 96.4% with immediate sealing and 81.8% without. That is one subgroup of one study: it supports the technique rather than promising it prevents debonding in every case.

One consequence is worth stating plainly, because patients often assume the opposite. Removing more tooth does not necessarily produce a more secure veneer, and past a point it can produce a less secure one, because it trades the surface that bonds well for the surface that bonds less well.

The word veneer covers several materials, and they do not all attach by the same chemistry. Glass-ceramics such as feldspathic porcelain and lithium disilicate are etched and then treated with a silane coupling agent, giving a bond that is both mechanical and chemical. Zirconia is not a glass, so etching does not produce the same surface and silane alone has little to hold on to: a durable bond generally needs the fitting surface mechanically prepared, by airborne-particle abrasion or silica coating, together with an adhesive containing the phosphate monomer MDP. A 2023 review of 93 laboratory studies found that combination gave the most durable bonds after artificial ageing. Those are laboratory measurements rather than clinical survival data, which is a real limitation, but the practical point holds: what a veneer is made of changes what a correct bonding procedure looks like.

An adhesive joint fails when the force across it exceeds what it can hold, so the second half of the question is never the adhesive on its own. A bite that distributes evenly spreads the work across many teeth. A bite that does not can concentrate it onto a few, and a veneered upper incisor taking a disproportionate share of the closing or sliding contact is holding against a force the design never accounted for. Clenching and grinding raise this considerably, applying greater force for far longer than chewing does, and if the occlusion was never recorded and checked, that pattern is invisible in a photograph and entirely visible on examination.

Underneath, two things undermine the joint rather than overload it. Decay at or under the margin removes the substrate the resin is holding, so the bond does not so much break as lose what it was attached to. An existing composite filling under the veneer is a third material in the stack with its own ageing behaviour, and bonding to it is not the same as bonding to tooth. Moisture matters as well, since adhesive bonding is intolerant of contamination during cementation, though that is the one thing nobody can reconstruct afterwards.

Recurrent debonding is multi-causal, so an assessment looks for which factor is operating in this mouth rather than assuming a universal cause. A useful examination establishes four things: what the restoration was bonded to, enamel or dentin or an existing filling; what it is made of and how its fitting surface was treated, since the material decides the protocol; whether decay or a failing older restoration is involved underneath; and how the tooth is loaded, with the contacts recorded rather than judged by eye and any history of grinding taken seriously.

Preparation and margin design belong in that assessment too. It is mechanically coherent that margin configuration and ceramic wrap influence how force crosses the joint, but the largest long-term series available tested preparation design directly and found no significant effect on survival, reporting the same for cement type. Worth examining, then, rather than treated as an established independent factor.