Merging an intraoral scan with a 3D face is clinically acceptable, but the evidence is weak
A systematic review of six studies concludes that merging a digital dental model with a 3D photogrammetric facial scan is clinically acceptable — most, though not all, outcomes stayed within a 2 mm or 2 degree threshold — with the certainty of evidence rated very low by GRADE.
An alternative to CBCT that still lacks evidence
Combining a digital dental model or intraoral scan with a three-dimensional photogrammetric facial scan offers orthodontics something appealing: a complete 3D representation of the relationship between teeth and soft tissue without a single radiograph. The question is how accurate that merge is. A systematic review in the European Journal of Orthodontics looked for an answer in MEDLINE, EMBASE, Web of Science and Cochrane, together with a hand search of reference lists, taking a full facial CBCT scan as the reference method.
The yield was modest: four eligible studies, plus two more after the search was updated. All confirmed that merging an intraoral scan with a 3D facial scan is feasible. All, however, were rated at high risk of bias with the QUADAS-2 tool, and the certainty of the evidence on accuracy is very low by GRADE. Because of differences in registration procedures and in the anatomical landmarks chosen, the authors could not pool the results and described them qualitatively instead.
The conclusion is conditionally positive: registration accuracy is clinically acceptable, and most — but not all — outcomes stayed within a 2 mm or 2 degree threshold. Whether they do depends on the anatomical region, the imaging method and the measurement method. The authors describe the method as a promising alternative without ionising radiation for orthodontic diagnosis and treatment planning, with the caveat that inconsistent registration procedures and small samples mean it needs further study. The review protocol was registered under the number INPLASY202420038.
Limitations
The review rests on six studies with a high risk of bias and small samples, and the heterogeneity of the procedures made pooling impossible — the verdict of clinically acceptable accuracy therefore rests on a qualitative synthesis, not on pooled figures.
For your practice
- If you already take intraoral scans and 3D facial photographs, merging them makes sense for patient communication and soft-tissue planning — not as a replacement for CBCT where CBCT is indicated.
- The 2 mm and 2 degree threshold is a working reference, but it does not hold equally across all facial regions; check how your software performs the registration.
- In children and adolescents the dose-avoidance argument is strongest, but while the evidence is this weak the decision on CBCT is still driven by clinical indication.