探讨早孕期(胎龄为11^＋0～13^＋6周)超声检测指标预测胎儿重型α地中海贫血的价值。

选择2015年1月3日至2016年12月30日，于广州市妇女儿童医疗中心接受定期产前检查、夫妻双方均为轻型α地中海贫血基因携带者所孕育的282例单胎妊娠胎儿为研究对象。对所有胎儿于胎龄为11^＋0～13^＋6周时进行超声筛查。早孕期超声检测指标包括：胎儿心胸直径比(CTR)、大脑中动脉血流峰值流速(MCA-PSV)和胎盘厚度(PT)。若胎儿的上述超声检测指标筛查结果呈阳性，则对其进一步进行绒毛穿刺或者羊膜腔穿刺，进行α地中海贫血基因检测；若呈阴性，则进行孕期随访，并于生后对新生儿常规进行脐带血血红蛋白(Hb)电泳分析，判断胎儿是否罹患重型α地中海贫血。根据产前α地中海贫血基因检测结果，或新生儿Hb电泳分析结果，将所有胎儿分别纳入重型组(重型α地中海贫血胎儿)与对照组(轻型α地中海贫血胎儿及健康胎儿)。采用成组t检验，对重型组与对照组胎儿早孕期超声筛查的CTR、MCA-PSV、PT值进行比较。绘制早孕期胎儿超声筛查的CTR、MCA-PSV、PT值，预测胎儿重型α地中海贫血的受试者工作特征(ROC)曲线，并计算ROC曲线下面积(ROC-AUC)，根据约登指数最大原则，确定早孕期胎儿超声筛查的CTR、MCA-PSV、PT值，预测胎儿重型α地中海贫血的最佳临界值，并计算其敏感度、特异度和似然比。本研究遵循的程序符合广州市妇女儿童医疗中心人体试验委员会制定的标准，并经过该伦理委员会批准(批准文号：2014120106)。

①根据282例胎儿的随访结果，最终61例(21.6%)被确诊为重型α地中海贫血，而被纳入重型组(n＝61)；221例为轻型α地中海贫血或健康胎儿，则被纳入对照组(n＝221)。2组胎儿胎龄构成比等一般临床资料比较，差异无统计学意义(P>0.05)。②重型组胎儿的早孕期CTR、MCA-PSV、PT值分别为(0.52±0.04)、(1.61±0.58) cm/s、(1.30±0.20) mm，均显著超过对照组的(0.44±0.03)、(1.20±0.30) cm/s 、(1.00±0.30) mm，2组比较，差异均有统计学意义(t＝19.470、9.320、6.700，P<0.001)。③ROC曲线分析结果显示，早孕期胎儿的CTR、MCA-PSV、PT值，对于预测胎儿重型α地中海贫血的ROC-AUC分别为0.949(95%CI：0.909～0.990，P<0.001)，0.811(95%CI：0.741～0.811，P<0.001)，0.778(95%CI：0.720～0.856，P<0.001)。根据约登指数最大原则，早孕期胎儿的CTR、MCA-PSV、PT值，对于预测胎儿重型α地中海贫血的最佳临界值分别为0.48、1.50 MoM及1.50 MoM(MoM为中位数倍数)，此时其预测胎儿重型α地中海贫血的敏感度分别为92.7%、48.2%及24.6%，特异度分别为95.5%、90.8%及94.1%，似然比分别为20.6、5.2及5.0。

早孕期胎儿超声检测相关指标，可有效预测重型和非重型α地中海贫血胎儿，其中预测敏感度和特异度最高的均为早孕期胎儿超声检测的CTR。对于重型α地中海贫血高危胎儿，若早孕期对胎儿的超声检测指标均为正常，则胎儿罹患重型α地中海贫血的几率低。

To evaluate the values of ultrasound testing markers for prediction of fetal severe α-thalassemia in the first trimester pregnancy(11^{+ 0}-13^{+ 6} weeks of gestational age).

From January 3, 2015 to December 30, 2016, a total of 282 cases of fetuses with singleton pregnancy whose parents both were mild α-thalassemia gene carriers and received regular prenatal examinations at the Guangzhou Women and Children′s Medical Center were selected as research subjects. All the fetuses received ultrasound screening in the first trimester pregnancy. And ultrasound testing markers including fetal cardiothoracic ratio (CTR), peak systolic velocity of middle cerebral artery (MCA-PSV), and placental thickness (PT) were prospectively assessed and recorded. For those with positive results of the above-mentioned ultrasound testing markers, chorionic puncture or amniocentesis was used to detect α-thalassemia gene and diagnose fetal severe α-thalassemia; for those with negative results, they were followed up until birth, and the umbilical cord blood hemoglobin (Hb) electrophoretic analysis was performed routinely after birth to diagnose whether the fetus was severe α-thalassemia. According to the α-thalassemia gene detection results or Hb electrophoretic analysis results, all fetuses were divided into severe group (homozygous α-thalassemia-1 fetus) and control group (mild α-thalassemia fetus and healthy fetus). The values of fetal CTR, MCA-PSV and PT in the first trimester pregnancy of severe group and control group were compared by independent-samples t test. The receiver operator characteristic (ROC) curves of fetal CTR, MCA-PSV and PT of ultrasound testing in the first trimester pregnancy for prediction of fetal severe α-thalassemia were drawn, and the area under ROC curve (ROC-AUC) were calculated. The optimal critical values of fetal CTR, MCA-PSV and PT in the first trimester pregnancy for prediction of fetal severe α-thalassemia were obtained when the Youden index reaching the maximum value. And their sensitivities, specificities and likelihood ratios were calculated. The procedures followed in this study were in accordance with the ethical standards established by the Human Subjects Trial Committee of Guangzhou Women and Children′s Medical Center, and this study was approved by this committee (Approval No. 2014120106).

①According to the follow-up results of 282 fetuses, 61 cases (21.6%) were diagnosed as severe α-thalassemia fetus and included in severe group (n=61), and 221 cases as mild α-thalassemia or healthy fetus were included in control group (n=221). There was no significant difference between the two groups in general clinical data such as the ratio of different gestational age (P>0.05). ②The CTR, MCA-PSV and PT values in the first trimester pregnancy of fetuses in severe group were (0.52±0.04), (1.61±0.58) cm/s, (1.30±0.20) mm respectively, which were significantly higher than those of control group (0.44±0.03), (1.20±0.30) cm/s, (1.00±0.30) mm, and there were significant differences between two groups (t=19.470, 9.320, 6.700; all P<0.001). ③ROC curve analysis showed that the ROC-AUC of fetal CTR, MCA-PSV and PT values in the first trimester pregnancy for prediction of fetal severe α-thalassemia were 0.949 (95%CI: 0.909-0.990, P<0.001), 0.811 (95%CI: 0.741-0.811, P<0.001), 0.778 (95%CI: 0.720-0.856, P<0.001) respectively. According to the maximum principle of Yoden index, the optimal cut-off values of CTR, MCA-PSV and PT in the first trimester pregnancy for prediction of fetal severe α-thalassemia were 0.48, 1.50 MoM and 1.50 MoM (multiples of the median), and the sensitivities for prediction of fetal severe α-thalassemia were 92.7%, 48.2% and 24.6% respectively, and the specificities were 95.5%, 90.8% and 94.1% respectively, and the likelihood ratios were 20.6, 5.2 and 5.0 respectively.

The ultrasound testing markers in the first trimester pregnancy can effectively predict fetal severe and non-severe α-thalassemia, and the most sensitive marker was fetal CTR. For fetuses with high-risk of severe α-thalassemia, if all the ultrasound testing markers are normal in the first trimester pregnancy, the probability of severe α-thalassemia is low.