Li Yahong, Sun Yun, Yang Lan, Huang Mingtao, Zhang Xiaojuan, Wang Xin, Guan Xianwei, Yang Peiying, Wang Yan, Meng Lulu, Zhou Ran, Zhou Xiaoyan, Luo Chunyu, Hu Ping, Jiang Tao, Xu Zhengfeng
Center of Prenatal Diagnosis, Women's Hospital of Nanjing Medical University, Nanjing Maternity and Child Health Care Hospital, Nanjing, China.
Department of Prenatal Diagnosis, Wuxi Maternity and Child Health Hospital Affiliated to Nanjing Medical University, Wuxi, China.
Front Cardiovasc Med. 2021 Jun 7;8:671191. doi: 10.3389/fcvm.2021.671191. eCollection 2021.
Congenital heart disease (CHD) is the most common birth defect. The prenatal diagnosis of fetal CHD is completely dependent on ultrasound testing, but only ~40% of CHD can be detected. The purpose of this study is to find good biomarkers in amniotic fluid (AF) to detect CHD in the second trimester, so as to better manage this group of people and reduce the harm of CHD to the fetus. Metabolites analysis were performed in two separate sets. The discovery set consisted of 18 CHD fetal maternal AF samples and 35 control samples, and the validation set consisted of 53 CHD fetal maternal AF samples and 114 control samples. Untargeted metabolite profiles were analyzed by gas chromatography/time-of-flight-mass spectrometry (GC-TOF/MS). Orthogonal partial least square discrimination analysis (OPLS-DA) demonstrated that CHD and control samples had significantly different metabolic profiles. Two metabolites, uric acid and proline, were significantly elevated in CHD and verified in two data sets. Uric acid was associated with CHD [odds ratio (OR): 7.69 (95% CI: 1.18-50.13) in the discovery set and 3.24 (95% CI:1.62-6.48) in the validation set]. Additionally, uric acid showed moderate predictive power; the area under curve (AUC) was 0.890 in the discovery set and 0.741 in the validation set. The sensitivity and specificity of uric acid to detect CHD was, respectively, 94.4 and 74.3% in the discovery set and 67.9 and 71.9% in the validation set. The identification of uric acid as a biomarker for CHD has the potential to stimulate research on the pathological mechanism of CHD and the development of a diagnostic test for clinical applications.
先天性心脏病（CHD）是最常见的出生缺陷。胎儿CHD的产前诊断完全依赖于超声检查，但只有约40%的CHD能够被检测出来。本研究的目的是在羊水（AF）中寻找良好的生物标志物，以在孕中期检测CHD，从而更好地管理这一群体并减少CHD对胎儿的危害。代谢物分析在两个独立的样本集中进行。发现集包括18例CHD胎儿的母体AF样本和35例对照样本，验证集包括53例CHD胎儿的母体AF样本和114例对照样本。采用气相色谱/飞行时间质谱（GC-TOF/MS）分析非靶向代谢物谱。正交偏最小二乘判别分析（OPLS-DA）表明，CHD样本和对照样本具有显著不同的代谢谱。两种代谢物，尿酸和脯氨酸，在CHD样本中显著升高，并在两个数据集中得到验证。尿酸与CHD相关[发现集中的比值比（OR）：7.69（95%CI：1.18-50.13），验证集中的比值比（OR）：3.24（95%CI：1.62-6.48）]。此外，尿酸显示出中等的预测能力；发现集中的曲线下面积（AUC）为0.890，验证集中的曲线下面积（AUC）为0.741。尿酸检测CHD的敏感性和特异性在发现集中分别为94.4%和74.3%，在验证集中分别为67.9%和71.9%。将尿酸鉴定为CHD的生物标志物有可能促进对CHD病理机制的研究以及临床应用诊断测试的开发。
