Precision of maxillo-mandibular registration with intraoral scanners in vitro

To compare the precision of maxillo-mandibular registration and resulting full arch occlusion produced by three intraoral scanners in vitro.

Six dental models (groups A-F) were scanned five times with intraoral scanners (CEREC, TRIOS, PLANMECA), producing both full arch and two buccal maxillo-mandibular scans. Total surface area of contact points (defined as regions within 0.1mm and all mesh penetrations) was measured, and the distances between four pairs of key points were compared, each two in the posterior and anterior.

Total surface area of contact points varied significantly among scanners across all groups. CEREC produced the smallest contact surface areas (5.7-25.3mm²), while PLANMECA tended to produce the largest areas in each group (22.2-60.2mm²). Precision of scanners, as measured by the 95% CI range, varied from 0.1-0.9mm for posterior key points. For anterior key points the 95% CI range was smaller, particularly when multiple posterior teeth were still present (0.04-0.42mm). With progressive loss of posterior units (groups D-F), differences in the anterior occlusion among scanners became significant in five out of six groups (D-F left canines and D, F right canines, p<0.05).

Maxillo-mandibular registrations from three intraoral scanners created significantly different surface areas of occlusal contact. Posterior occlusions revealed lower precision for all scanners than anterior. CEREC tended towards incorrect posterior open bites, whilst TRIOS was most consistent in reproducing occluding units.