Conformity of Virtual Surgical Plans to Postoperative Results in Segmental Orthognathic Surgery

Abstract

Background/purposeVirtual surgical planning (VSP) in orthognathic surgery has enabled reconstructive surgeons to achieve more predictive results when compared to conventional methods. Still, few studies have evaluated the accuracy of VSP in more complex cases, such as those requiring maxillary segmentation. The aim of this study was to examine the conformity of virtual surgical plans to postoperative results following segmental procedures performed by a single surgeon. MethodsThis was a retrospective study of patients who had undergone triple jaw (segmental Le Fort I, bilateral sagittal split osteotomy, and genioplasty by the senior author (DMS) between 2016 and 2020. The preoperative virtual surgical plan and postoperative cone beam computed tomographic scan were registered in 3-Matic software (Materialise, Leuven, Belgium) using cranial base landmarks. The conformity of the postoperative skull model to the virtual surgical plan was quantified using linear and angular measurements at various 3D bony landmarks. The average mean difference (AMD) in the x-, y-, and z-axes was calculated. Paired T-tests were used to compare postoperative measurements with those from the pre-operative STL. Multivariate linear regression was performed to analyze and modifying effects of age, gender, number of segments (2 vs. 3), and malocclusion class. ResultsTwenty-nine patients were included (mean age 24.5 years; 72.4% female). Three-dimensional analysis showed significant AMD between the virtual plan and postop CBCT for the following landmarks: U1 Midpoint (y, AMD: 1.65mm, P = .03), B point (y, AMD: 1.63mm, P = .005), pogonion (y, AMD: 2.38mm, P = .0005), menton (y, AMD: 2.22mm, P = .0003). There were no differences between the virtual plans and postop CBCT at any other points. Transversely, the overall AMD was 1.15mm, with under-expansion occurring in 79.3% (2.18mm) and over-expansion in 20.7% of patients (1.65mm). Angular measurements revealed statistical significance for SNB (AMD: 1.38, P = .02), but not for SNA and ANB values. Multivariate regression showed no association between age, gender, segmentation (2 vs. 3), or class of malocclusion and the extent of conformity for any measurements. ConclusionThere is a high degree of conformity between the VSP and postoperative results following triple jaw orthognathic surgery with segmental maxillary osteotomies. The greatest nonconformity was found in the sagittal plane and was mostly limited to the mandible. Segmentation did not affect surgical accuracy of the maxillary movements.