RESUMO
OBJECTIVE: Cephalometric inclination change of the lower incisors during orthodontics is used to assess treatment outcome. The lower border of the mandible is commonly used for measuring inclination change, despite it being subject to remodelling in growing patients. Superimposition of radiographs using Björk's 'stable structures' is intended to exclude these growth changes. We tested whether there is a significant difference for three commonly used methods to assess inclination change induced by orthodontic treatment (Me-Go, Go-Gn, the tangent to the lower border of the mandible) when comparing it to Björk's 'stable structures'. METHODS: Björk's superimposition does not allow measuring incisor inclination changes directly; hence, one pre- and mid-treatment cephalogram of 39 growing orthodontic patients were superimposed in this retrospective study. The radiographs were taken at least 1 year apart (120 weeks; SD = 34.4). Patients undergoing growth modification treatment were excluded. Standardized cephalograms were hand traced and changes in lower incisor inclination, using the three mandibular planes, were compared to the changes obtained by anatomical superimposition of Björk's 'stable structures'. RESULTS: Linear regression showed good intra-class correlation (ICC) between all methods. ICC was 0.96 for Me-Go, 0.94 for Go-Gn, and 0.92 for the lower border tangent. ICC for operator reliability was 0.99. LIMITATIONS: Measurement errors affect all investigations of both analogue and digital radiographs, but movement artefacts particularly apply to the latter. Cephalometry uses two-dimensional measurements of a three-dimensional subject, which can lead to further inaccuracies. These limitations have to be taken into account when interpreting the results of our investigation. CONCLUSION: Data obtained from Björk's superimposition did not vary significantly from the other more commonly used techniques (Me-Go, Go-Gn, and the tangent to the lower border of the mandible). Remodelling of the lower border of the mandible was insignificant for the time period investigated.