妊娠特异性甲状腺激素参考值范围制定及妊娠期合并亚临床甲状腺功能减退预测模型的建立  

Establishment of Trimester-specific Thyroid Hormones Reference Ranges and Construction of a Prediction Model for Subclinical Hypothyroidism During Pregnancy

作　　者：罗丽娜 火睿[1] 李爱玲[3] 庞磊[4] 孙侃[1] 常向云[1] 赵霞[2] 刘红梅[2] 吕晨 李辰璐 刘嘉奇 陈静[1] LUO Lina;HUO Rui;LI Ailing(Department of Endocrinology,The First Affilited Hospital of Shihezi University,Shihezi Xinjiang 832000,China;Obstetrics Department of Shihezi City People's Hospital,Shihezi Xinjang 832000,China)

机构地区：[1]石河子大学第一附属医院内分泌科,新疆石河子832000 [2]石河子大学第一附属医院产科,新疆石河子832000 [3]石河子市人民医院产科,新疆石河子832000 [4]石河子市妇幼保健院妇女保健科,新疆石河子832000

出　　处：《实用妇产科杂志》2023年第4期308-313,共6页Journal of Practical Obstetrics and Gynecology

基　　金：石河子大学科技计划项目(编号:ZZZC201820A);石河子大学高层次人才科研启动项目(编号:RCZX201536);八师石河子市社会发展科技攻关与成果转化项目(编号:2018YL03)。

摘　　要：目的:建立石河子地区妊娠期妇女甲状腺激素参考值范围,筛选孕妇发生亚临床甲状腺功能减退(SCH)的独立危险因素,构建预测模型并验证。方法:采用回顾性队列研究分析2020年6月至2021年6月石河子3家围产门诊常规产检的孕妇,收集患者的一般资料,检查血清甲状腺激素、血常规、生化、凝血全项等指标,建立本地区妊娠各期甲状腺激素参考值范围,并计算甲状腺疾病的患病率。通过Logistic回归分析筛选SCH相关危险因素,构建预测模型。采用受试者工作特征(ROC)的曲线下面积(AUC)、校准曲线及决策曲线分析(DCA)验证与评价该模型的区分度、校准度和临床实用价值。结果:(1)妊娠期促甲状腺激素(TSH)水平随孕周的增加而递增,游离甲状腺素(FT4)则递减,差异有统计学意义(P<0.001)。(2)SCH是妊娠期最常见的甲状腺疾病,其次为单纯甲状腺自身抗体阳性和单纯低甲状腺素血症。(3)二元Logistic回归结果显示,叶酸(OR0.98,95%CI0.97~0.98,P<0.001)、维生素B_(12)(OR0.99,95%CI 0.98~0.99,P<0.001)、血红蛋白(OR0.98,95%CI 0.97~0.99,P=0.033)水平升高是SCH的保护性因素,而谷氨酰转肽酶(OR1.01,95%CI1.01~1.02,P<0.001)、胆固醇(OR2.11,95%CI 1.60~2.77,P<0.001)、白细胞(OR1.05,95%CI 1.02~1.08,P=0.003)、嗜酸性粒细胞(OR12.55,95%CI 2.50~63.06,P=0.002)水平升高是SCH发生的独立危险因素。(4)以上述7个因素构建列线图模型并绘制ROC曲线,AUC为0.81(95%CI 0.77~0.85,P<0.001),经过内部验证后模型C指数为0.80。DCA显示模型具有良好的临床应用价值。结论:本研究构建的列线图模型可个体化预测孕妇发生SCH风险,准确度和临床实用性较高。Objective:To establish the normal reference ranges of thyroid hormones in each trimester of preg-nancy in Shihezi region,screen the independent risk factors of subclinical hypothyroidism(SCH),and then con-struct and verity the prediction model of SCH.Methods:Pregnant women with routine prenatal testing in three perinatal clinics from June 2020 to June 2021 were included,serum thyroid hormone,blood routine,serum bio-chemistry,blood coagulation,and other indicators were tested and recorded,trimester-specific thyroid hormones reference ranges were established,and prevalence rate of thyroid disease was calculated.SCH-related risk fac-tors were screened by Logistic regression analysis to construct prediction models.Receiver operating characteristic curve(ROC),calibration curve and decision curve analysis(DCA)were used to verify and evaluate the differential validity,calibration accuracy and clinical utility value of the model.Results:①The level of thyroid stimulating hormone(TSH)was increased with gestational weeks,while free thyroxine(FT4)decreased(P<0.001).②SCH.was the most common thyroid disease in pregnancy,followed by euthyroid women with thyroid autoantibody posi-tivity and isolated hypothyroxinemia.③Binary Logistic regression showed that high levels of folic acid(OR 0.98,95%CI0.97-0.98,P<0.001),vitamin B12(OR0.99,95%CI0.98-0.99,P<0.001),and hemoglobin(OR 0.98,95%CI 0.97-0.99,P=0.033)were protective factors for prevention of SCH,while the high levels of glu-tamyl transpeptidase(OR 1.01,95%CI 1.01-1.02,P<0.001),cholesterol(OR2.11,95%CI 1.60-2.77,P<0.001),white blood cells(OR1.05,95%CI 1.02-1.08,P=0.003),and eosinophils(OR 12.55,95%CI2.50-63.06,P=0.002)were independent risk factors for the occurrence of SCH.④The Nomogram model was con-structed based on above factors,with an AUC of 0.81(95%CIO.77-0.85,P<0.001)and a model C index of 0.80 after internal validation,showed a good clinical application value according to the DCA.Conclusions:In this study,the nomogram model can individually predict SCH risk in pre

关 键 词：亚临床甲状腺功能减退 妊娠 参考值 预测模型 

分 类 号：R714.256[医药卫生—妇产科学]