孕早期促甲状腺素和β-人绒毛膜促性腺激素联合预测子痫前期的应用价值

Value of thyroid stimulating hormone combined with β-human chorionic gonadotropin in the prediction of preeclampsia in the first trimester of pregnancy

目的评价在孕早期应用促甲状腺素(thyroid stimulating hormone,TSH)和β-人绒毛膜促性腺激素(β-human chorionic gonadotropin,β-h CG)联合预测子痫前期(preeclampsia,PE)的诊断价值。方法对有早期产检并分娩的368例产妇的资料进行回顾性队列研究分析,根据是否发生PE分为PE组(52例)和对照组(316例),分析患者一般资料、甲状腺功能、β-h CG以及妊娠相关血浆蛋白A(pregnancy-associated plasma proteins A,PAPP-A)等指标,并以临床诊断作为"金标准",应用受试者工作特征(receiver operating characteristic,ROC)曲线比较不同指标对于预测PE的诊断价值以及诊断效能。结果单因素分析显示,PE组产妇的孕前体质量指数[BMI,(24.41±2.11) kg/m^2]、TSH[(2.79±0.83) mIU/L]、PAPP-A[(1.27±0.28) mIU/L)]水平明显高于对照组[(23.72±1.96) kg/m^2,(1.99±0.53) mIU/L,(0.98±0.39) mIU/L](P=0.021、0.000、0.000),而β-hCG水平[(3.06±1.72)×10~4 IU/L]显著低于对照组[(5.48±2.29)×10~4 IU/L](P=0.000)。Logistic回归分析显示,TSH(OR=15.423,95%CI=12.963～59.162,P=0.000)和PAPP-A(OR=9.794,95%CI=4.195～22.867,P=0.000)是PE的独立危险因素,β-hCG(OR=0.889,95%CI=0.853～0.927,P=0.000)是PE的保护因素。ROC曲线显示,TSH和β-hCG预测PE的曲线下面积(AUC=0.780,0.788)高于PAPP-A(AUC=0.705)和孕前BMI(AUC=0.575),Youden指数得出TSH和β-hCG的最佳截点分别为TSH≥2.54 mIU/L和β-hCG≤3.96×10~4 IU/L;应用TSH联合β-hCG(AUC=0.927)的诊断价值明显高于两者单独预测。TSH联合β-hCG预测PE的敏感度为96.15%,均显著高于TSH的84.62%和β-hCG的82.69%(x^2=3.983、4.981;P=0.046、0.026),而诊断准确率差异均无统计学意义(x^2=0.363、3.160;P=0.547、0.075)。结论在孕早期应用TSH和β-hCG联合预测PE具有较高的应用价值,其中TSH≥2.54 mIU/L和β-hCG≤3.96×10~4 IU/L是较好的截点。

Objective To observe the value of thyroid stimulating hormone (TSH) combined with β-human chorionic gonadotropin (β-hCG) in the prediction of preeclampsia (PE) in the first trimester of pregnancy.Methods A retrospectively cohort analysis was performed on 368 pregnant women who had undergone early clinical examination and delivery. The patients were divided into PE group (52 cases) and control group (316 cases) according to whether PE occurred. The general data, thyroid function, β-hCG and pregnancy-associated plasma protein A (PAPP-A) were comparatively analyzed. Using the clinical diagnosis as the "golden standard" , the receiver operating characteristic (ROC) curves were used to compare the diagnostic value and diagnostic efficacy of those indicators for predicting adverse pregnancy outcomes.Results Univariate analysis showed that pre-pregnancy body mass index [BMI, (24.41±2.11) kg/m^2], TSH [(2.79±0.83) mIU/L] and PAPP-A [(1.27±0.28) mIU/L)] levels in PE group were significantly higher than those in control group [(23.72±1.96) kg/m^2, (1.99±0.53) mIU/L, (0.98±0.39) mIU/L](P=0.021, 0.000, 0.000), while β-hCG level [(3.06±1.72)×10^4 IU/L] was significantly lower than that in control group [(5.48±2.29)×10^4 IU/L](P=0.000). Multivariate logistic regression analysis showed that TSH (OR=15.423, 95% CI=12.963~59.162, P=0.000) and PAPP-A (OR=9.794, 95% CI=4.195~22.867, P=0.000) were independent risk factors for PE, while β-hCG (OR=0.889, 95% CI=0.853~0.927, P=0.000) was a protective factor of PE. ROC curves showed that the area under the curve of TSH and β-hCG (AUC=0.780, 0.788) was higher than that of PAPP-A (AUC=0.705) and pre-pregnancy BMI (AUC=0.575), the Youden index showed that the best cut-off of TSH and β-hCG were TSH≥2.21 mIU/L and β-hCG≤ 3.96×10^4 IU/L. The diagnostic value of TSH combined with β-hCG (AUC=0.927) was significantly higher than that of TSH or β-hCG alone. The sensitivity of TSH combined with β-hCG in predicting PE was 96.15%, which was significantly higher than that of TSH and β-hCG (χ^2=3.983, 4.981;P=0.046, 0.026), while no statistical differences were found in the diagnostic accuracy (χ^2=0.363, 3.160;P=0.547, 0.075).Conclusion TSH combined with β-hCG have a high value in the prediction of PE in the first trimester of pregnancy, of which the best cut-off of TSH and β-hCG are TSH≥2.21 mIU/L and β-hCG≤3.96×10^4 IU/L, respectively.