| Objective: Symmetry is essential for a normal, healthy person. Asymmetry anisotropy is typical in craniomaxillofacial deformities, which seriously distorts patients' figure and psychology. Classical plastic and reconstructive surgery has not been very effective in correcting serious asymmetrical deformity in the past due to lack of precision. This study aims to improve this precision by using digital craniomaxillofacial plastic and reconstructive surgery based on principals of reversed engineering. The study used fresh pigs' heads as samples. Experimental processes included collecting CT data, CAD/CAM, rapid prototyping (RP), and computer-aided milling. This produced skull molds of the pigs' heads and individual implant pieces. Then, measurements were taken from the skull entities, CT images, 3-D reconstruction images and the RP resin model, and the measurements were compared to find the error generated within each stage, as well as the overall difference between the final skull molds and the skull entities. Finally the technique was applied in clinic to repair craniofacial defects and clinical effects were observed.Methods: Part 1 was animal experiments. Experimental objects were 4 fresh pigs' heads. Three groups of symmetry markers were chosen for measurements on the pigs' heads, and were identified by inserting titanium screws into the marking points. Then, the pigs' heads were scanned with CT. The CT data were written on a disc in DICOM format. After this step, skull specimen of the pigs' heads were made. Experiment 1: CT data were imported into 3-D MSR software to perform a visible 3-D surface reconstruction. The 3-D reconstruction images were exported to rapidform in .stl format for RP resin model manufacture. The distance between the symmetry markers were measured and results were compared among the skull entities, CT data, 3-D reconstruction images and the resin model. Experiment 2: A designed region from one side of zygomatic arch was cutoff from the pig's skull specimen. A CT scan was made to the remaining part of the skull specimen, and CT data were imported into 3-D MSR for a 3-D surface reconstruction. Export the reconstruction image into Surfacer9.0 software to perform a virtual osteotomy of the same region. The digital piece was then sent for rapid prototyping and computer-aided milling respectively for manufacture of resin model and titanium implant. At last, the piece of zygomatic arch and both resin model and titanium implant were measured and compared statistically to find the differences between manufacture methods and entity.Part 2 was clinical practice. Seven patients suffered craniomaxillofacial defects were successfully repaired with customized digital implants using above mentioned techniques through proper surgical approach.Results: The results of animal experiments and clinical application showed that the resin model of skull made by digital engineering technique duplicated its entity very well in structure. The customized digital implants made with mirror technique adapted to the surface of the defect very well. The results of symmetrical repair and reconstruction for asymmetrical deformity of craniomaxillofacial region were satisfactory.1 Through measuring the distances between the symmetry markers on skull entities, CT images, 3-D reconstruction images and the RP resin model, the result of analysis of variance did not show significant differences among four groups (F=0.021, P=0.996>0.05).2 Compared to the skull entity, the errors in distance between symmetrical markers were as follows: 1.54±1.31mm for CT images, 2.62±1.64mm for 3-D reconstruction images, 1.89±0.79mm for RP resin models.3 For RP resin models compared to the entity, average error in distance between symmetrical markers was 1.89±0.79mm (0.09~3.03mm), and average error in distance between symmetrical marker and the middle line was 0.94±0.40mm (0.05~1.52mm).4 In comparison of the measurements among the zygomatic arch, the RP resin model and the titanium implant, analysis of variance among the three groups did not show significant differences (F=0.497,P=0.615>0.05). 5 The error between titanium implant and skull entity was 1.16±0.39mm, the error between RP model and skull entity was 1.17±0.47mm, the error between titanium implant and the RP resin model was 0.00±0.26mm.6 In clinical application, the surgical approach must be firstly taken into account because of irregular shape of customized implant. Then the digital implant will be fixed to the defect very easily. The implant adapts the surface of the defect very well. The operative time is significantly shortened, the surgical procedure simplified, the trauma of operation decreased, the symmetry of patient's appearance satisfied. During 2 years follow-up, no obvious side effects and dysfunctions were found.Conclusion1 The RP skull model can reflect three-dimensional anatomy of the skeleton intuitively and in detail. Customized implant made by digital engineering adapts the surface of defect and the result of surgery was satisfied.2 The errors between RP skull model and titanium implant to the skull entity arise mainly from treatment process of the image. In these processes, CT data extraction and the surface treatment after visible three-dimensional reconstruction are the main reason for the errors.3 During 2 years follow-up, the results of digital craniomaxillofacial plastic and reconstructive surgery were satisfactory, which has simplified the operation, shortened the operative time, and decreased the operative trauma.4 Although the error between digital titanium implant and skull entity can be visibly neglected in clinic, there is discordance of augmentative quantity between hard and soft tissues. It needs further investigation whether or not to compensate the customized implant in the process of design and manufacture.
|
PDF Full Text Request