Intraoral scanners: trueness and precision are different things, and the one that matters is the second

Intraoral scanner datasheets quote an accuracy figure in micrometres, and that number is read as though it described the quality of the instrument. It describes only half the problem.

The ISO standard distinguishes two quantities. Trueness measures how close the scan comes to reality; precision measures how closely two scans of the same object resemble each other.

A scanner can be highly precise and poorly true: it repeats the same error every time. On a single unit this matters little, because the restoration is built on that scan. On a full arch, trueness becomes decisive, because errors accumulate along the path.

This explains the most debated behaviour of the technology: scanners perform very well over short distances and lose reliability across full arches.

The reason lies in how they build the model. Each frame is aligned to the previous one, and every alignment introduces a minimal error. Over forty millimetres the accumulated error is negligible; across a whole arch it becomes measurable.

Systematic reviews report deviations under 50 micrometres on single units and short spans, with values that can exceed 100 micrometres on full arches, and marked differences between systems.

Scanning strategy influences the result more than manufacturers' documentation suggests. A path that retraces its own steps creates additional reference points and reduces drift; a linear scan from one end to the other maximises it.

The absence of reference points is the physical limit. An extended edentulous area, smooth and without recognisable anatomy, offers the scanner nothing on which to align frames, and that is where error grows.

It is also why intraoral scanning of fully edentulous arches remains less reliable than conventional impression, except where artificial reference markers are applied to the mucosa.

Moisture behaves differently from what one might expect. Scanners not requiring opaque powder tolerate saliva well but not moving liquid: a wet, still surface scans; a running droplet produces artefacts.

The real advantage is not accuracy but workflow. No disinfection, no shipping, no distortion in transit, and the ability to review the preparation on an enlarged screen and correct it in the same appointment.

That immediate check is often worth more than the difference in micrometres: an incomplete margin seen on screen is corrected in two minutes; the same margin discovered in the laboratory costs an appointment.

In summary: single units and short bridges are scanner territory, where trueness and precision are both adequate. Full arches call for case-by-case judgement, and in complete edentulism the conventional impression retains a documented advantage.