[Multivariate analysis of mandibular morphology with fetal growth by means of the soft standardized radiographs].

Mandibular growth of the fetus is a great matter of importance to anatomy, anthropology and also oral and Maxillofacial surgery. The postnatal changes of the mandible that have occurred with growth have been reported by numerous investigators. However, a few researchers have investigated the prenatal growth of the mandible. The purposes of this study were discussed how to change in size and shape of the mandible during the fetal period. The materials consisted of 162 human fetuses from the body length (BL) of 95 mm, to 500 mm. They were divided into four groups: 95-199 mm (BL1, N = 40), 200-299 mm (BL2, N = 76), 300-399 mm (BL3, N = 26), 400-500 mm (BL4, N = 20). All specimens were stored in formalin, and only those free from gross deformation were selected for study. The heads were removed by decapitation at the level of the hyoid bone. The soft standardized radiographs were taken of the frontal, lateral, basal aspects. They were used for tracing and angular and linear measurements (total thirty). The data were used to evaluate the mandibular form by multivariate analysis. They were supported the univariate analysis. The findings leads to the following conclusions: 1) Cluster analysis and principal component analysis were applied to the data to make clear the correlation between 30 variables. Four clusters were constructed at the level ninety similarities in cluster analysis. They were corresponded to four factors were obtained in principal component analysis. 2) The data were subjected to a stepwise discriminant function analysis using the seventeen of original 30 variables. Nine variables, mandibular length (Go-Me), mandibular ramus height (Col-Go), Symphyseal height (Id-Me), mandibular body height (Kr'-beta), gonial angle (< Co2, Go, Me), mental angle (< Id, Pog, Go), gonial width (BGo), alveolar width (BKr'), basal angle of mandible (< RGo, Me, LGo), were selected as the best discriminators. The incidence of correctly classified cases 88.68%. 3) As a result of the first discriminant function of the discriminant analysis and one-way analysis of variance (ANOVA) and t-test, mandibular length (Go-Me), gonial width (BGo), alveolar width (BKr'), Symphyseal height (Id-Me) were longer in tall group that short group (BL4 > BL3 > BL2 > BL1). Therefore, the first discriminant function was accounted for the size factor. 4) In consequence of the second discriminant function of the discriminant analysis and regression analysis, gonial width (BGo) was grown broader than alveolar width (BKr') from BL1 to BL2.(ABSTRACT TRUNCATED AT 400 WORDS)