Hannan W J, Cowen S J, Wrate R M, Barton J
Department of Medical Physics and Medical Engineering, Western General Hospital, Edinburgh.
Arch Dis Child. 1995 Feb;72(2):147-9. doi: 10.1136/adc.72.2.147.
To establish improved predictive values for normal bone mineral content (BMC) and density (BMD) in adolescent girls a community based study of schoolgirl volunteers was carried out by dual energy x ray absorptimetry. Measurements were performed on 216 subjects aged 11.0 to 17.9 years; measurements were repeated one year later on 84 of the girls, providing a total of 300 studies. For total body BMC the standard error of the prediction was improved from 290.9 g to 134.1 g when weight, height, and shoulder width were added to the normal variable of age. For spine BMD the standard error of the prediction was improved from 0.105 g/cm2 to 0.066 g/cm2 when height, weight, and shoulder width were added to the normal variable of age. Significant improvements were also obtained for total BMD and spine BMC. Despite the normal practice of predicting bone density from age alone this was not selected as the first variable in a multiple stepwise regression for either total body or spine. The prediction of BMC and BMD can be significantly improved by the inclusion of simple body habitus parameters in the prediction equations. As the mean (SE) z score derived from the manufacturer's normal data was - 0.36 (0.02) these American data are probably not appropriate for use in adolescents in the UK.
为了建立青春期女孩正常骨矿物质含量(BMC)和骨密度(BMD)的改进预测值,采用双能X线吸收法对在校女志愿者进行了一项基于社区的研究。对216名年龄在11.0至17.9岁的受试者进行了测量;一年后对其中84名女孩重复测量,共进行了300次研究。对于全身BMC,当将体重、身高和肩宽添加到年龄这一常规变量中时,预测的标准误差从290.9克降至134.1克。对于脊柱BMD,当将身高、体重和肩宽添加到年龄这一常规变量中时,预测的标准误差从0.105克/平方厘米降至0.066克/平方厘米。全身BMD和脊柱BMC也有显著改善。尽管通常仅根据年龄来预测骨密度,但在全身或脊柱的多步回归中,年龄并非被选为第一个变量。通过在预测方程中纳入简单的身体形态参数,BMC和BMD的预测可以得到显著改善。由于根据制造商的正常数据得出的平均(标准误)z分数为 -0.36(0.02),这些美国数据可能不适用于英国的青少年。