Error analysis of a CAD/CAM method for unidirectional mandibular distraction osteogenesis in the treatment of hemifacial microsomia.

Our aim was to investigate the errors in a computer-aided design and manufacture (CAD/CAM) method of unidirectional mandibular distraction osteogenesis. Six patients with hemifacial microsomia were selected, and studied on computed tomographic (CT) scans taken at 3 time intervals: preoperatively, at the end of the latent period, and at the end of consolidation. The plan for mandibular distraction osteogeneisis was designed using CT-based 3-dimensional visible software. The osteotomy line and site of the drill were transferred to a rapid prototyping surgical guide. The osteotomy of the mandible and implantation of the distraction device were completed under guidance. The accuracy of the transferred surgical plan was confirmed by fusion of images after the latency period. The 3-dimensional superimposition of the preoperative simulation, and the postoperative actual models at the end of consolidation, showed that the mean (SD) error between the actual and the predicted height of the ramus was 0.6 (0.6) mm. The error between the actual and predicted intercondylar distance was 8.1 (2.1) mm. There was a significant difference in intercondylar distance between the simulated and actual groups (p=0.00024). The 3-dimensional CT-based planning system described in this paper was transferred precisely from the virtual plan to the real-time operation. The planning system also gave a precise prediction of the height of the ramus after mandibular distraction osteogenesis. However, because of the pull of the lateral pterygoid muscle and pseudarthrosis, the intercondylar distance decreased compared with the predicted value. These influencing factors should be considered when the planning system is refined.