Orbital Floor Reconstruction With Patient-Specific Implants: Comparable Outcomes Between Mirroring and Virtual Reduction Techniques

Background Orbital floor fractures may cause enophthalmos, diplopia, and facial asymmetry, often requiring surgical intervention to restore form and function. Advances in virtual planning enable customized reconstruction; however, the influence of specific planning techniques on surgical accuracy remains uncertain. Purpose This study aimed to compare the accuracy of patient-specific implant (PSI) placement, using 2 the virtual reconstruction techniques: mirroring and virtual fracture reduction. Study Design, Setting, and Sample This retrospective cohort study was conducted at the Galilee Medical Center (Nahariya, Israel). Adults with isolated orbital floor fractures treated between 2021 and 2024 were included. Patients with bilateral fractures, prior orbital surgery, or incomplete data were excluded. Predictor Variable The predictor variable was the virtual surgical planning technique: mirroring or virtual fracture reduction. Main Outcome Variables The primary outcome was PSI accuracy, defined as root mean square deviation (mm) between planned and postoperative implant positions. Accuracy between techniques was considered statistically equivalent when no significant difference was found ( P > .05). Secondary outcomes included orbital volume symmetry, enophthalmos, and diplopia. Covariates Covariates were subject age, sex, affected side, and follow-up duration. Analyses Due to non-normal distribution, Mann–Whitney U and Wilcoxon signed-rank tests were applied, with significance set at P < .05. Results The study included 23 subjects (mean age, 43.1 ± 15.7 years), of whom 14 (61%) were male. Postoperative root mean square deviation between the planned and actual PSI positions was 0.67 ± 0.22 mm in the mirroring group and 0.62 ± 0.24 mm in the virtual reduction group ( P = .6). Orbital volume improved from 108.6 to 99.5% with mirroring and from 107.3 to 101.1% with virtual reduction ( P < .05 within groups; P > .05 between groups). Preoperative diplopia occurred in 38 and 44% of subjects in the 2 groups, respectively, and resolved in most cases. Enophthalmos was present in over 80% preoperatively and improved in both groups. Conclusions and Relevance The comparable results between the 2 methods support the flexibility of using either approach based on case specifics and the designer's preference, experience, and familiarity with the planning technique.