Abstract: Objective To know the distribution of TSH and incidence of neonatal congenital hypothyroidism(CH),and study the screening cut-off points of the thyroid-stimulating hormone(TSH)in Anqing. Methods The neonatal screening data about 59 295 newborns in Anqing were analyzed and the TSH cut-off points were calculated by method of percentiles and Receiver Operating Characteristic (ROC)curve,then,the incidence of sufferers was analyzed by different cut-off points. Results The distribution of the neonatal screening data was deflection,P₅₀ was 2.52 mU/L,P₉₉ was 8.94 mU/L.Combined with ROC curve,the cut-off points of our lab was 8.12 mU/L. Conclusions Different cut-off points effect the consequences of the neonatal screening.It is important to establish the suitable cut-off points with ROC curve based on lab.

Key words: congenital hypothyroidism, thyrotropic-stimulating hormone, neonatal screening, ROC curve, cut-off points

摘要： 目的 探讨制定合理的本地新生儿促甲状腺素(TSH)筛查切值。方法 对本市59 295例新生儿TSH筛查结果进行分析,使用百分位数法及ROC曲线法制定TSH筛查切值。结果 新生儿TSH水平呈偏态分布,P₂₅为1.47 mU/L,P₅₀为2.52 mU/L,P₇₅为3.96 mU/L,P₉₅为6.89 mU/L ,P₉₉为8.94 mU/L;ROC曲线法得出本实验室TSH筛查切值为8.12 mU/L。按ROC筛查切值诊断22例先天性甲状腺功能减退症(CH)和28例高TSH血症,若按P₉₉和试剂盒给定的切值标准将漏诊CH患儿1例,漏诊率达4.55%。结论 选用ROC曲线法确定适宜的TSH切值是提高原发性先天性甲状腺功能减退症筛查质量的重要措施。

关键词: 先天性甲状腺功能减退症, 促甲状腺素, 筛查, ROC曲线, 切值