Leurs I H, Wattel E, Aartman I H A, Etty E, Prahl-Andersen B
Department of Orthodontics, Academic Centre for Dentistry Amsterdam, Louwesweg 1, 1066 EA Amsterdam, The Netherlands.
Eur J Orthod. 2005 Jun;27(3):309-14. doi: 10.1093/ejo/cji010.
Dental age was studied in a sample of 451 Dutch children (226 boys and 225 girls) according to the method of Demirjian. They were born between 1972 and 1993 and were between 3 and 17 years of age at the time a dental pantomogram (DPT) was obtained. All children were placed in the age group closest to their chronological age. All 451 DPTs were scored by one examiner. A subset of 52 DPTs was scored by a second examiner and the intra-class correlation coefficient (ICC) and Cohen's kappa were calculated. The ICC was 0.99 and Cohen's kappa 0.68. Boys and girls were analysed separately.A significant difference was found between chronological age and dental age. On average, the Dutch boys were 0.4 years and the girls 0.6 years ahead of the French-Canadian children analysed by Demirjian. Therefore, the French-Canadian standards were not considered suitable for Dutch children. New graphs for the Dutch population were constructed using a logistic curve with the equation Y = 100*{1/(1 + e(-alpha(x - x0)))} as a basis. The 90 per cent confidence interval was calculated. To determine whether the logistic curve was correct, a residual analysis was carried out and scatter plots of the differences were made. The explained variance was 93.9 per cent for the boys and 94.8 per cent for the girls. Both the residual analysis and the scatter plots indicated that the logistic curve was appropriate for use with Dutch children. In addition to the graphs, tables were produced which transfer the maturity scores calculated by the method of Demirjian into Dutch dental age.
根据德米尔坚(Demirjian)方法,对451名荷兰儿童(226名男孩和225名女孩)的样本进行了牙龄研究。他们出生于1972年至1993年之间,在获得口腔全景X线片(DPT)时年龄在3至17岁之间。所有儿童都被归入与其实际年龄最接近的年龄组。所有451张DPT均由一名检查者评分。52张DPT的子集由另一名检查者评分,并计算组内相关系数(ICC)和科恩kappa系数。ICC为0.99,科恩kappa系数为0.68。对男孩和女孩分别进行了分析。发现实际年龄和牙龄之间存在显著差异。平均而言,荷兰男孩比德米尔坚分析的法裔加拿大儿童超前0.4岁,女孩超前0.6岁。因此,法裔加拿大标准被认为不适用于荷兰儿童。以方程Y = 100*{1/(1 + e^(-alpha(x - x0)))}的逻辑曲线为基础,为荷兰人群构建了新的图表。计算了90%的置信区间。为了确定逻辑曲线是否正确,进行了残差分析并制作了差异散点图。男孩的解释方差为93.9%,女孩为94.8%。残差分析和散点图均表明逻辑曲线适用于荷兰儿童。除了图表之外,还制作了表格,将通过德米尔坚方法计算的成熟度得分转换为荷兰牙龄。