根据采血时间调整新生儿先天性甲状腺功能减退症筛查切值的研究分析

Study on Adjusting the Screening Cut-Off Value of Neonatal Congenital Hypothyroidism According to Blood Collection Time

摘要: 目的：探讨根据采血时间调整新生儿先天性甲状腺功能减退症筛查切值的研究；方法：回顾分析2018年~2019年在我院产三病区出生的并进行促甲状腺素初筛的新生儿5144例，按照患儿的采血日龄分为7 d以下和7 d及以上，将2018~2019年的筛查结果作为促甲状腺素初筛切直的调整依据，2020年筛查结果用来验证调整后切值的准确性切值的特性，采用ROC曲线图和百分位数法进行确定，计算切值的特异度、敏感度以及阳性和阴性预测值；结果：2018~2019年共筛查新生儿5144例，初筛查阳性28例，复筛查阳性5例，其中3例确诊，除此之外，有2例新生儿在初筛查时结果显示为阴性，进过复查之后确诊为甲状腺功能减退症，最终确诊甲状腺功能减退症人数为3例，2020年共筛查新生儿5432例，初筛查阳性42例，复筛查阳性13例，其中10例确诊，除此之外，有5例新生儿在初筛查时结果显示为阴性，经过复查之后确诊为甲状腺功能减退症，最终确诊甲状腺功能减退症人数为5例，其确诊率为1/1440 (95% CI: 1/1325~1/1743)，对5144例新生儿的促甲状腺素初筛浓度用不同切值进行特异度敏感度、阳性和阴性预测值进行评估和研究，随着设定切值的逐渐提高，评估的敏感度也逐渐下降且特异度呈逐渐升高趋势，当促甲状腺素切值设定为8 mIU/L时约登指数最大，且敏感度为98.2%，特异度98.7%，ROC曲线下面积0.906，按调整前和调整后的切值对2020年新生儿促甲状腺素筛查结果进行比较：调整后阳性例数由3587例降至2169例，从调整前的1.26%降至0.76%，阳性率差异有统计学意义(P < 0.01)，阳性预测值从调整前的6.04%增至7.86%。结论：根据采血时间调整促甲状腺素切值能够很大程度地减少假阳性的情况发生，值得临床推广。

Abstract: Objective: To explore the study of adjusting the screening cut-off value of neonatal congenital hy-pothyroidism according to the blood collection time; Methods: 5144 newborns who were born in the three wards of our hospital from 2018 to 2019 and underwent thyrotropin primary screening were analyzed retrospectively. They were divided into less than 7 d and more than 7 d according to the age of blood collection. The screening results from 2018 to 2019 were used as the adjustment basis for thyrotropin primary screening and straightening, the screening results in 2020 are used to ver-ify the accuracy of the adjusted cut-off value. The characteristics of the cut-off value are determined by the ROC curve and percentile method, and the specificity, sensitivity, positive and negative pre-dictive values of the cut-off value are calculated; Results: from 2018 to 2019, a total of 5144 new-borns were screened, 28 were positive in the initial screening and 5 were positive in the re-screening, of which 3 were diagnosed. In addition, 2 newborns showed negative results in the in-itial screening and were diagnosed with hypothyroidism after re-examination. The number of final-ly diagnosed hypothyroidism was 3. In 2020, a total of 5432 newborns were screened and 42 were positive in the initial screening, 13 cases were positive in re-screening, of which 10 cases were con-firmed. In addition, 5 newborns showed negative results in the initial screening. After re-examination, they were diagnosed with hypothyroidism. The number of finally diagnosed hypo-thyroidism was 5, and the diagnosis rate was 1/1440 (95% CI: 1/1325~1/1743), the specificity, sensitivity, positive and negative predictive values of thyrotropin in 5144 newborns were evaluated and studied with different cut-off values. With the gradual increase of the set cut-off value, the sen-sitivity of the evaluation decreased and the specificity increased gradually. When the cut-off value of thyrotropin was set to 8 mIU /L, the yoden index was the largest, with the sensitivity of 98.2% and specificity of 98.7%, the area under the ROC curve was 0.906. The results of neonatal thyrotropin screening in 2020 were compared according to the cut-off values before and after adjustment: the number of positive cases after adjustment decreased from 3587 to 2169, from 1.26% to 0.76%, the difference in positive rate was statistically significant (P < 0.01), and the positive predictive value increased from 6.04% to 7.86%. Conclusion: Adjusting thyrotropin cut-off value according to blood collection time can greatly reduce the occurrence of false positive, which is worthy of clinical promotion.