Facial Asymmetry Evaluation Post Mandibular Reconstruction Using Whole Versus Segmented Mandibular In-House 3D Model Guides: A Randomized Clinical Trial.

INTRODUCTION: Three-dimensional (3D) printing enhances surgical precision in mandibular reconstruction by providing patient-specific models. This study compares the use of a 3D-printed whole-mandible model versus a segmented mandible as guides for mandibular reconstruction using a free fibular flap (FFF) to achieve a symmetrical facial outcome. OBJECTIVES: To compare clinical and radiologic facial asymmetry after mandibular reconstruction using FFF guided by whole-mandible versus segmented mandible models in-house 3D-printed surgical guides. METHODS: This randomized controlled trial included patients undergoing segmental mandibular reconstruction with FFF between June 2023 and June 2025. Participants were randomly assigned to reconstruction guided by either whole-mandible or segmented mandible in-house 3D-printed models. Models were generated from preoperative CT data using open-access software and used intraoperatively to guide fibular osteotomy and plate contouring. Postoperative facial asymmetry was evaluated at 1 month using standardized frontal photographs and anteroposterior head radiographs, quantified by facial asymmetry indices based on established cephalometric landmarks. Outcomes were compared between groups using appropriate parametric or nonparametric statistical tests in (IBM SPSS Statistics Software version 29.0; IBM Corp., Armonk, NY). RESULTS: A total of 36 patients were enrolled and randomized equally to the whole-mandible and segmented-mandible groups; 2 postoperative dropouts were excluded from the outcome analysis. Baseline characteristics, defect classification, and the number of fibular segments were comparable between groups (P>0.05). On clinical assessment, the segmented group demonstrated lower facial asymmetry indices at the sella-nasion and mandibular angles than the whole-mandible group, although the differences were not statistically significant (P>0.05). Radiologic evaluation showed similar trends, with no significant differences between groups (P>0.05). CONCLUSION: The segmented mandible model showed slightly better outcomes, though not statistically significant, while the whole-mandible model remained reliable when intraoperative modification was required. Further studies on virtual surgical planning are needed to support cost and efficiency analyses.