Accuracy of virtual patient reconstruction for complete arch implant-supported prostheses with extraoral scan body-assisted registration.

STATEMENT OF PROBLEM Extraoral scan bodies have been increasingly used to reconstruct virtual patients for complete arch implant-supported prostheses. However, their accuracy compared with that of conventional tooth surface-guided registration for reconstructing the jaw-to-bone positional relationship of virtual patients remains unknown.

Purpose

The purpose of this study was to compare the accuracy and repeatability of 2 methods of reconstructing the jaw-to-bone positional relationships in virtual patients for complete arch implant-supported prostheses: scan body-assisted registration and tooth surface-guided registration. MATERIAL AND

Methods

Facial scans, prosthesis scans, and cone beam computed tomography (CBCT) data from 22 participants were collected to create virtual casts using tooth surface-guided and scan body-assisted registration (test groups) and CBCT reconstruction (control). Deviations in jaw-to-face measurements from the control were calculated for each test group and then compared using paired t tests or the Wilcoxon signed-rank test (α=.05). Repeatability was evaluated by performing each registration twice and comparing intramethod deviations.

Results

The scan body-assisted registration group showed significantly lower mean measurement deviation than the tooth surface-guided group (1.89 ±0.74 mm compared with 2.34 ±0.75 mm, P=.006). After correction for multiple comparisons, a statistically significant difference between the 2 groups was found in the linear distance between the subnasale (SN) and the anterior nasal spine (ANS). Both methods demonstrated comparable repeatability.

Conclusions

Extraoral scan body-assisted registration demonstrated significantly higher accuracy than tooth surface-guided registration in reconstructing the jaw-to-face positional relationship of a virtual patient for a complete arch implant-supported prosthesis. Both registration methods exhibited good repeatability.