RESUMO
The aim of this study was to verify the reproducibility and accuracy of preoperative planning in maxilla repositioning surgery performed with the use of computer-aided design/manufacturing technologies and mixed reality surgical navigation, using new registration markers and the HoloLens headset. Eighteen patients with a mean age of 26.0 years were included. Postoperative evaluations were conducted by comparing the preoperative virtual operation three-dimensional image (Tv) with the 1-month postoperative computed tomography image (T1). The three-dimensional surface analysis errors ranged from 79.9% to 97.1%, with an average error of 90.3%. In the point-based analysis, the errors at each point on the XYZ axes were calculated for Tv and T1 in all cases. The median signed value deviation of all calculated points on the XYZ axes was -0.03mm (range -2.93mm to 3.93mm). The median absolute value deviation of all calculated points on the XYZ axes was 0.38mm (range 0mm to 3.93mm). There were no statistically significant differences between any of the points on any of the axes. These values indicate that the method used was able to reproduce the maxilla position with high accuracy.