Validation of Andrews Analysis in the Virtual Environment

Abstract

Andrews analysis is a tool to establish the aesthetic anteroposterior position of the maxilla. Andrews analysis has not been evaluated through computer-aided surgical simulation (CASS). The purpose of this study was to evaluate the accuracy of Andrews profile analysis when performed in the virtual environment. A retrospective cohort study was implemented with consecutive patients undergoing orthognathic surgery between February 2020 and February 2022 at the University of Alabama, Birmingham. Traditional Andrews analysis with lateral smiling photographs were taken during the presurgical appointment in adjusted natural head position (aNHP). The standard cone-beam CT obtained for CASS and archived on the KLS Martin (Jacksonville, Florida) database was accessed for retrospective measurement. Lateral facial photographs in aNHP were imported into the virtual environment and the three dimensional (3D) composite model was then oriented into aNHP. The software engineer, blinded to the traditional measurements, then performed the Andrews analysis in the virtual environment by placing a vertical glabella line on the 3D composite model in aNHP. The linear horizontal distance of the maxillary central incisor perpendicular to the vertical glabella line was recorded. Method of Andrews analysis measurement (traditional photographic evaluation vs CASS) MAIN OUTCOME VARIABLE: Linear Andrews analysis measurement. Additional covariates evaluated were sex, age at surgery, and dentofacial deformity diagnosis. Descriptive statistics were computed to compare photographic analysis versus CASS analysis. A P value of <.05 was considered statistically significant. The average age was 25.7years old and 54% of patients were female. For the photographic analysis, the mean incisor-goal anterior limit line distance was -0.44±7.12mm (95% CI, -1.13 to 0.37mm; P=.46). For the virtual analysis, the mean incisor-goal anterior limit line distance was 0.13±7.21 (95% CI,-0.004 to 0.30; P=.89). The Pearson correlation coefficients between the photograph and 3D analysis were very strong (0.93). The root mean square deviation between the photographic and 3D analysis cohorts was 2.7mm. Given the high correlation coefficients between all demographics, CASS can be utilized for Andrews analysis to determine ideal anteroposterior maxillary position to streamline data collection and the planning process.