When a restoration fails: the cause almost always lies upstream of the symptom

Remaking a failed restoration with the same technique that produced it yields the same result, some years later. Diagnosing the cause is the part of the work most often skipped, because the patient wants a solution, not an analysis.

Repeated debonding is almost always a retention problem, not a cement problem. A short or overly tapered abutment does not hold, and changing cement postpones the failure by a few months without resolving it.

The check is simple: measure abutment height and assess convergence. Below 3 millimetres of height or above 20 degrees of total taper, that is the cause and must be corrected with additional retentive features or crown lengthening.

If instead the geometry is adequate, examine the occlusal contacts. Debonding on a retentive preparation almost always indicates an eccentric contact generating tensile forces during lateral movement.

Ceramic fracture has different causes depending on where it occurs. Fracture at the incisal edge in a patient with steep anterior guidance indicates functional overload; fracture in the posterior occlusal area more often indicates insufficient thickness.

Fracture involving the underlying framework is instead a design problem: a connector too thin in a bridge, or an excessive cantilever. Remaking the same design produces the same outcome.

Recurrent fractures in the same location in a bruxist patient have a cause no material resolves. There the choice lies between a tougher monolithic material and a nightguard, and generally both are needed.

Fracture of the core beneath a crown is the worst outcome because it often means extraction. The commonest cause is absence of the ferrule effect: without a band of sound circumferential dentine, load concentrates at the base of the build-up.

When a fractured core is found, before rebuilding one must verify whether sufficient ferrule exists for the new crown. If not, repeating the work means scheduling the next failure.

Marginal leakage with secondary caries indicates an unsealed margin, and the causes divide in two. A fit defect in the crown, detectable with a probe; or a margin positioned in an area the patient cannot clean.

The second is more frequent and is a design error: a deep subgingival margin in a patient with difficult hygiene produces caries regardless of the restoration's accuracy.

Failure analysis requires examining the removed piece, not only the tooth. The fracture pattern, presence of residual cement and wear on the occlusal surfaces tell what happened better than any patient account.

In summary: repeated debonding means retention or occlusion, ceramic fracture means thickness or load, core fracture means absent ferrule, leakage means a badly positioned margin. Every failure indicates a precise correction, and repeating the work without making it is the one certainly wrong choice.