Comparison of 2 methods of making surgical models for correction of facial asymmetry.

PURPOSE Stereolithography is useful in reconstructive surgery in that the surgical template or customized implant can be prefabricated on the models. To correct facial asymmetry, prior reshaping of the replica of the original structures is frequently required before it can be used as a surgical model. This is traditionally accomplished by direct sculpturing. This method has its limitations in clinical use. Recently, we developed a method using computer techniques to reconstruct the required structures. We herein report several of its applications in a variety of clinical situations and compare this virtual method with the traditional method. PATIENTS AND METHODS With the traditional method, reconstruction of the models was handmade on the replica of the original structures. In the virtual method, the anticipated reconstructions were completed on the computer using various image-processing tools and were verified by the surgeons before sending to stereolithography. Thirteen patients who had undergone surgical correction of facial asymmetry using models made by either method were retrospectively reviewed. The traditional method was used in 5 of them while the virtual method was applied in the other 8 patients. The surgical models and the patients following the reconstruction were evaluated for symmetry and esthetics. RESULTS To construct implants or to precontour fixation plates, an average of 1.4 models was fabricated for each patient using the traditional method, whereas only 1.1 models were made for each patient in the virtual method group. Both methods worked satisfactorily in restoring symmetry of the bony structures on the models. However, the projection of the chin on the model created by the traditional method was inadequate, as showed postoperatively in 1 patient. There was surface roughness on the customized area of the models made by the virtual method. The surgical result was poor in symmetry in 1 case in the traditional method group. One patient in the virtual method group showed irregularities on the temporal region following augmentation with prefabricated bone cement implant. CONCLUSIONS Both methods of making models were useful and effective in surgical reconstruction for facial symmetry in selected cases. The virtual method was preferred in cases where midline structures had already been deformed, or when soft tissue was involved in reconstruction. From the technical standpoint, the virtual method was superior because of its versatility, predictability, precision, communicability, and the convenience of storage and documentation.