An in vitro accuracy study on scan body-assisted surface based registration for conventional and zygomatic dental implants.

There are various techniques to measure accuracy of dental implant surgery, but limited data validating the techniques used to analyze accuracy. Scan-body-assisted surface-based registration (SB-SBR) is deemed theoretically accurate, but with challenges in testing accuracy. The purpose of the study was to analyze the accuracy of SB-SBR for conventional (CVI) and zygomatic implants (ZI) using 2 complementary techniques. An in-vitro study was designed using a 3D-printed model of an edentulous maxilla with CVI and ZI digitally planned and placed. SB-SBR was performed on the physical model and on a model-free virtual setting, and the implant position was compared with those of the planned implants 10 times. Outcome variables were angular and linear errors (AE and LE), with thresholds of 1° and 1 mm. Paired t-test and Wilcoxson signed-ranked test were used. Results showed greater AE for ZI versus CVI (mean difference, MD 0.42°). Apical LE was greater for ZI (MD 0.68 mm), while CVI exhibited greater platform LE (MD 0.69 mm). Overall AE remained under 1°, but the apical LE of ZI exceeded 1 mm, which could potentially lead to violation of critical structures when combined with surgical error, highlighting the need for caution when interpreting data derived from SB-SBR especially for ZI.