Accuracy of splint vs splintless technique for virtually planned orthognathic surgery: A voxel-based three-dimensional analysis

Abstract

Purpose: This study compared two transferring methods for virtually planned orthognathic surgery - the CAD/CAM intermediate splint and the customized surgical guide with fixation plates.

Methods: This was a prospective clinical study in which participants were consecutively recruited and underwent bimaxillary orthognathic surgery. They were divided into two groups based on the transferring method used. The pre- and postoperative CBCTs were aligned using voxel-based landmark-free registration, and the discrepancies for selected points were compared with the planned displacement of the virtually planned surgery. The maxilla and mandible were analyzed separately, and translation and rotation movements were considered.

Results: A total of 16 patients, divided into two groups of eight patients each, were included in this study. The splintless group was significantly more accurate for the translation movement along the x-axes for points A (p = 0.008; mean absolute error 0.527 ± 0.387 for the splint group and 0.137 ± 0.067 for the splintless group) and Ans (p = 0.045; mean absolute error 0.535 ± 0.446 for the splint group and 0.156 ± 0.002 for the splintless group). For the mandible there was a significant difference in accuracy along the x-axes for points B (p = 0.049; mean absolute errors 1.728 ± 1.181 and 0.697 ± 0.519 for the splint and splintless groups, respectively), LL3 (p = 0.049; mean absolute error 1.629 ± 0.912 and 0.851 ± 0.797 for the splint and splintless groups, respectively), LR3 (p = 0.049; mean absolute error 1.711 ± 0.906 and 0.844 ± 0.780 for the splint and splintless groups, respectively), with the splintless group being more accurate. For the rotation the splintless group was significantly more accurate along the y-axes (p = 0.04; mean absolute error 1.62 ± 0.78 and 0.49 ± 0.31 for the splint and splintless groups, respectively) and z-axes (p = 0.04; mean absolute error 0.63 ± 0.45 and 0.17 ± 0.05 for the splint and splintless groups, respectively) for the maxilla, while no significant difference was found for the mandible.

Conclusions: Overall, the customized fixation plate system is more accurate than the intermediate CAD/CAM splint for transferring the virtual plan into the operation room.

Keywords: Accuracy; Orthognathic surgery; Three-dimensional evaluation; Virtual planning; Voxel-based landmark-free registration.