The aim of this study was to determine a differential diagnostic model for anterior open bite (AOB) in children in order to predict evolution with growth in cases of skeletal or dysfunctional dysplasia.

The sample was composed of 128 subjects divided into 2 groups according to overbite values: a control group with normal occlusion and 2.5±1mm overbite, and an AOB group with a negative overbite. Measurements of hyoid bone position, gonial angle, SN/Go-Gn, FMA, height of alveolar processes, upper lip height, height of the anterior nasal spine to stomion superior, upper incisor and menton, antegonial notch, symphysis anatomy, condyle anatomy, convexity, facial axis, lower oropharyngeal area and curve of Spee were obtained for all subjects, based on lateral cephalometric radiographs and lateral photos. All data were analyzed, and multivariate logistic regression was applied. The area under the receiver operating characteristic (ROC) curve was calculated for the models obtained by logistic regression, in order to evaluate them. Statistical analysis was performed in IBM^® SPSS^® Statistics version 21.0.

The results demonstrated that the AOB model was capable of predicting skeletal anterior open bite with an accuracy of 93%. The seven variables selected were: lower face height (LFH), height of mandibular alveolar process, antegonial notch, symphysis direction, symphysis thickness, gonial angle and facial axis.

Measurements of facial morphology made it possible to construct a model able to differentiate, with a small margin of error, between the two distinct forms of AOB: skeletal and dysfunctional.