Three-dimensional analysis of mandibular landmarks, planes and shape, and the symphyseal changes associated with growth and orthodontic treatment
Deller, Cecilia Mercedes
MetadataShow full item record
OBJECTIVE: To test reliability of 3D mandibular landmarks, planes of reference and surfaces and assess their correlation to conventional 2D cephalometric measurements. To analyze changes in three-dimensional shape of the symphysis due to growth and orthodontic treatment. METHODS: This was a retrospective analysis of CBCTs of healthy orthodontic patients. 32 subjects were included, 16 males and 16 females. Mean ages of 10.6 ± 1.5 years and 15.0 ± 0.9 years before and after treatment, respectively. The mean follow up time was 4.3 years. Subjects free of any craniofacial anomalies, and no observable pathology on panoramic radiograph were. 15 subjects had CVM 1 and 17 subjects had CVM 2 before orthodontic treatment. All subjects had CVM 5 after orthodontic treatment. For the first phase, 3D mandibular landmark identifications were digitized. Planes and landmarks were constructed and compared with conventional 2D mandibular measurements. For the second phase, mandibles were isolated by removing surrounding structures. Pearson correlation and paired t-test were performed to test for correlation and differences between 2D and 3D measurements, respectively. Statistical analysis was performed using SAS 9.4. Software. MorphoJ software (Version 2.0, www.ﬂywings.org.uk) was used for symphysis shape analysis; and Discriminant Function Analysis (DFA) between pre-treatment and post-treatment was used for statistical analysis of the symphysis. RESULTS: We found statistical significant positive correlation between 2D and 3D pre-treatment ramus height (P-value =0.01), post-treatment ramus height (P-value < 0.0001), pre-treatment corpus length (P-value 0.0003), post-treatment corpus length (P-value 0.04), pre-treatment gonial angle (P-value <0.0001), and post-treatment gonial angle (P-value=0.05). Also, statistically significant differences in 2D ramus height (P=0.001), 3D ramus height (P-value=0.002), 2D corpus length (P-value <0.01), and 3D corpus length (P-value <0.01). For symphysis shape comparing between pre-treatment and post-treatment, we found that there is no statistically significant difference between them (P-value= 0.99). CONCLUSION: These results demonstrated statistically significant positive correlation between certain 2D and 3D measurements, pre-treatment and post-treatment differences in 2D and 3D measurements showed consistent results. Symphysis shapes do break out as distinctly separate groups, but the differences between the means is small.