G-P法与RUS-CHN法预测正常儿童成年身高的准确性研究

Accuracy of G-P Method and RUS-CHN Method in Predicting Adult Height of Normal Children

摘要: 目的：探讨G-P图谱法与RUS-CHN法预测正常儿童成年身高的准确性，为生长发育相关疾病的诊断及治疗提供依据。方法：回顾性分析100例正常儿童的骨龄片使用G-P图谱法与RUS-CHN法评估骨龄，用B-P法与中华05法预测身高，采用二元logistics回归对其骨龄与生活年龄、预测身高与实际成年身高进行相关分析，绘制ROC曲线并计算其曲线下面积，采用t检验对骨龄与实际年龄的差值、预测身高与成年身高的差值进行比较。结果：G-P图谱法及RUS-CHN法的预测身高和成年真实终身高均具有明显相关性(P < 0.05)。CHN-RUS法预测法的曲线下面积(0.726)较G-P图谱法(0.684)大，CHN-RUS法预测身高较G-P法性高；G-P图谱法的骨龄较实际年龄大(0.3226 ± 1.4009)，而RUS-CHN法骨龄较实际年龄约小(−0.1354 ± 1.1563)，两种方法的比较具有统计学意义(P < 0.05)；B-P法预测得到的平均身高比实际成年终身高高(1.46 ± 6.3917)，变异较大，中华05法预测身高较实际成年终身高矮(−1.0878 ± 4.9800)。结论：本研究两种骨龄评估方法及其相对应的身高预测法均适用于正常学龄儿童，且RUS-CHN法评估骨龄较G-P图谱法更接近生活年龄且预测身高准确性更高。

Abstract: Objective: To explore the accuracy of G-P map method and RUS-CHN method in predicting the adult height of normal children, so as to provide a basis for the diagnosis and treatment of growth-related diseases. Methods: The bone age films of 100 normal children were retrospectively analyzed using G-P atlas method and RUS-CHN method to evaluate bone age, using B-P method and Zhonghua 05 method to predict height, and using binary logistic regression to compare bone age and life age, predicted height and actual height. Correlation analysis was carried out for adult height, ROC curve was drawn and the area under the curve was calculated, and the difference between bone age and actual age and the difference between predicted height and adult height were compared by t test. Results: There was a significant correlation between the predicted height of G-P map method and RUS-CHN method and the true lifetime height of adults (P < 0.05). The area under the curve (0.726) of the CHN-RUS method was larger than that of the G-P atlas method (0.684), and the height pre-dicted by the CHN-RUS method was higher than that of the G-P method; The bone age of G-P atlas method is older than the actual age (0.3226 ± 1.4009), the bone age of the RUS-CHN method is about smaller than the actual age (−0.1354 ± 1.1563), and the comparison between the two meth-ods is statistically significant (P < 0.05); the average height predicted by the B-P method is higher than the actual adult lifetime height (1.46 ± 6.3917), the variation is large, the height predicted by the Zhonghua 05 method is higher than the actual adult lifetime height (−1.0878 ± 4.9800). Conclu-sion: In this study, the two bone age assessment methods and their corresponding height prediction methods are suitable for normal school-age children, and the RUS-CHN method is closer to the chronological age than the G-P atlas method and has higher accuracy in predicting height.