探讨无创产前检测(NIPT)联合孕妇血清可溶性转铁蛋白受体(sTfR)、生长分化因子(GDF)-15、红细胞体积分布宽度变异系数(RDW-CV)检测，对胎儿β-地中海贫血的诊断价值。

选择2017年1月至2018年6月，于自贡市妇幼保健院接受羊膜腔穿刺术确诊胎儿为β-地中海贫血的120例孕妇为研究对象。根据其妊娠胎儿的β-地中海贫血轻、中、重型，将其分别纳入轻型组(n＝48)、中型组(n＝40)与重型组(n＝32)。选取同期于本院建卡进行常规产前检查，并且妊娠胎儿发育正常的100例健康孕妇作为对照，纳入对照组。采集4组孕妇空腹肘静脉血，提取孕妇外周血浆中胎儿细胞外游离DNA(cfDNA)，采用2次PCR方法扩增后，采用反向斑点杂交方法检测胎儿β-地中海贫血基因。采用酶联免疫吸附测定法，测定4组孕妇血清sTfR、GDF-15、RDW-CV水平。对4组孕妇血清sTfR、GDF-15与RDW-CV水平分别总体比较，采用单因素方差分析，进一步两两比较，采用最小显著性差异(LSD)-t检验。与胎儿β-地中海贫血诊断"金标准"比较，计算NIPT联合孕妇血清sTfR、GDF-15、RDW-CV水平及这4项指标单独检测，对于胎儿β-地中海贫血诊断的敏感度、特异度、粗符合率、阳性预测值、阴性预测值。本研究严格按照2013年修订的《世界医学协会赫尔辛基宣言》要求进行。4组孕妇年龄及其妊娠胎儿的胎龄等一般临床资料比较，差异均无统计学意义(P>0.05)。

①夫妻双方β-地中海贫血基因型相同的40例孕妇中，NIPT与羊膜腔穿刺术对于胎儿β-地中海贫血诊断的符合率为90.0%(36/40)；夫妻双方β-地中海贫血基因型不同的80例孕妇中，上述2种方法对于胎儿β-地中海贫血诊断的符合率亦为90. 0%(72/80)。②4组孕妇血清sTfR、GDF-15、RDW-CV水平分别总体比较，差异均有统计学意义(F＝2 283.445、323.181、45.900，P均<0.001)，进一步两两比较，任意2组孕妇血清sTfR、GDF-15、RDW-CV水平比较，差异均有统计学意义(P<0.05)，并且孕妇血清sTfR、GDF-15、RDW-CV水平随着β-地中海贫血胎儿严重程度增加而逐渐升高。③NIPT联合孕妇血清sTfR、GDF-15、RDW-CV水平检测，对于胎儿β-地中海贫血诊断的敏感度、特异度、粗符合率、阳性预测值与阴性预测值，分别为86.7%、88.0%、87.3%、86.7%与84.6%，其特异度与阳性预测值，均显著高于单独检测孕妇血清sTfR、GDF-15、RDW-CV水平，并且差异均有统计学意义(与孕妇血清sTfR水平比较：χ²＝8.866、5.301，P＝0.003、0.021；与孕妇血清GDF-15水平比较：χ²＝9.765、5.929，P＝0.002、0.015；与孕妇血清RDW-CV水平比较：χ²＝7.167、4.327，P＝0.007、0.038)。

孕妇血清sTfR、GDF-15、RDW-CV可能成为诊断胎儿β-地中海贫血的生物标志物，NIPT联合这3项指标检测，可提高诊断胎儿β-地中海贫血的特异度与阳性预测值。由于本研究纳入样本量相对较小，NIPT联合孕妇血清sTfR、GDF-15、RDW-CV水平，对胎儿β-地中海贫血的诊断价值，仍然有待进一步研究、证实。

To investigate the diagnostic values of noninvasive prenatal testing (NIPT) combined with serum soluble transferrin receptor (sTfR), growth differentiation factor (GDF)-15, and red blood cell volume distribution coefficient of variation (RDW-CV) in pregnant women for diagnosis of fetal β-Thalassemia.

From January 2017 to June 2018, a total of 120 pregnant women whose fetuses were diagnosed as β-Thalassemia by amniocentesis in Zigong Maternal and Child Health Hospital were selected as research subjects. They were included into mild group (n=48), intermediate group (n=40) and severe group (n=32), respectively, according to mild, intermediate and severe β-Thalassemia of their fetuses. And other 100 cases healthy pregnant women who received routinely prenatally examination in the same hospital during the same period and with normal development of fetuses were selected as controls, and were included into control group. Fasting elbow venous blood was collected from 4 groups of fetal pregnant women, and fetal cell-free DNA (cfDNA) was collected from peripheral plasma of pregnant women. After amplification of cfDNA by twice PCR methods, fetal β-Thalassemia gene was detected by reverse dot blot hybridization. Enzyme-linked immunosorbent assay method was used to detect serum levels of sTfR, GDF-15 and RDW-CV of pregnant women in 4 groups. Serum levels of sTfR, GDF-15 and RDW-CV of pregnant women in 4 groups were compared by one-way ANOVA, and further multiple comparisons were performed by least significant difference (LSD)-t test. The sensitivity, specificity, crude accuracy, positive predictive value, and negative predictive value of combination of NIPT and serum sTfR, GDF-15 and RDW-CV levels of pregnant women, and detection of these four indicators alone for the diagnosis of fetal β-Thalassemia were calculated by comparison with " gold standard" of fetal β-Thalassemia. The procedures in this study were in line with the requirements of World Medical Association Declaration of Helsinki revised in 2013. There were no statistical differences in general clinical data such as fetal gestational age and age of pregnant women among four groups (P>0.05).

①Among 40 pregnant women who had the same genotype of β-Thalassemia with her husband, diagnosis accordance rate of NIPT and prenatal diagnosis amniocentesis for diagnosis of fetal β-Thalassemia was 90.0% (36/40). And among 80 pregnant women who had different β-Thalassemia genotype with her husband, diagnosis accordance rate of these two methods was 90.0% (72/80). ②The serum levels of sTfR, GDF-15, and RDW-CV of pregnant women in four groups were compared, and all the differences were statistically significant (F=2 283.445, 323.181, 45.900, Pall<0.001). Results of further multiple comparisons showed that there were statistically significant differences in serum levels of sTfR, GDF-15, and RDW-CV of pregnant women between each two groups (P<0.05), and serum levels of sTfR, GDF-15, and RDW-CV of pregnant women gradually increased with the increase of fetal β-Thalassemia severity. ③The sensitivity, specificity, crude accuracy, positive predictive value, and negative predictive value for diagnosis of fetal β-Thalassemia by NIPT combined with serum levels of sTfR, GDF-15, and RDW-CV of pregnant women were 86.7%, 88.0%, 87.3%, 86.7% and 84.6%, respectively. And its specificity and positive predictive value were significantly higher than those of serum sTfR, GDF-15, and RDW-CV levels detection alone in pregnant women, and all the differences were statistically significant (vs serum sTfR level detection in pregnant women: χ²=8.866, 5.301, P=0.003, 0.021; vs serum GDF-15 level detection in pregnant women: χ²=9.765, 5.929, P=0.002, 0.015; vs serum RDW-CV level detection in pregnant women: χ²=7.167, 4.327, P=0.007, 0.038).

Serum level of sTfR, GDF-15 and RDW-CV of pregnant women are expected to be the diagnostic markers of fetal β-Thalassemia. NIPT combined with detection of these three indicators can improve the specificity and positive predictive value for diagnosis of fetal β-Thalassemia. As the sample size in this study is relatively small, the diagnostic value of NIPT combined with serum levels of sTfR, GDF-15 and RDW-CV of pregnant women for fetal β-Thalassemia remains to be further studied and confirmed.