A method of comparing virtual reality orthognathic surgical predictions and postsurgical treatment outcomes

Background Orthognathic surgical predictions using 2D cephalometric films, 2D computer-assisted planning, and 3D computer-assisted surgical simulation are currently accepted methods to evaluate postsurgical outcomes. Recent advancements have paved the way for a new method of surgical prediction using immersive virtual reality. The objective of the study was to evaluate a method of comparing virtual reality (VR) orthognathic surgical predictions with postsurgical treatment outcomes of patients. Methods Pre- and postsurgical cone-beam computed tomography (CBCT) data of 14 patients who underwent one-jaw mandibular advancement surgery were collected. The presurgical CBCTs were rendered for the VR surgical manipulation, and the prediction was exported as a STL file. Skeletal and dental landmarks were placed, and differences were calculated in all dimensions (Right/Left, Anterior/Posterior, Superior/Inferior, 3D distance, pitch, roll, yaw). A one-sample t-test was conducted to determine clinical significance ( p < 0.05). Results The differences in the linear measurements of component dimensions for VR surgical prediction and actual postsurgical results were less than 2 mm ( p < 0.05), with the exceptions of left gonial angle in the right/left dimension; right and left gonial angles, left mental foramen, and gnathion in the anterior/posterior dimension; and all 3D distance measurements. The difference in all angular measurements (pitch, roll, yaw) were less than 4° ( p < 0.05). Conclusions The described method was successful in evaluating clinically acceptable limits in 3D distance measurements, component ( x , y , z ) dimensions and for pitch, yaw and roll of virtual reality surgical predictions and the actual surgical outcome.