Hobbs Charlotte A, MacLeod Stewart L, Jill James S, Cleves Mario A
University of Arkansas for Medical Sciences, College of Medicine, Department of Pediatrics, Arkansas Children's Hospital Research Institute, Little Rock, AR, USA.
Birth Defects Res A Clin Mol Teratol. 2011 Apr;91(4):195-203. doi: 10.1002/bdra.20784. Epub 2011 Mar 7.
The purpose of this study was to identify metabolic, genetic, and lifestyle factors that discriminate between women who have pregnancies affected by congenital heart defects (CHDs) from those who have unaffected pregnancies.
We analyzed the concentrations of 13 serum biomarkers, 3 functional genetic variants, and 4 lifestyle factors among 417 women with CHD-affected pregnancies and 250 controls. To identify risk factors that discriminated between cases and controls, we used logistic regression followed by recursive partitioning to identify non-linear interactions. A receiver operating characteristic (ROC) curve was constructed to evaluate the discriminatory accuracy of the final model.
A combination of risk factors discriminated women who had pregnancies affected by CHDs from those who had unaffected pregnancies. Among 21 possible determinants, serum concentrations of homocysteine and methionine, and reduced:oxidized glutathione ratios (GSH:GSSG) had the greatest discriminatory power. Recursive partition modeling resulted in five terminal nodes each illustrating the interplay of these three biomarkers. Women with elevated homocysteine and low GSH:GSSG had the highest risk of having CHD-affected pregnancy, whereas women with low homocysteine, high methionine, and high GSH:GSSG had the lowest risk. The corresponding area under the ROC curve was 81.6% (95% confidence interval [CI], 78.1-85.2%), indicating high ability to discriminate between cases and controls.
High homocysteine, low methionine, and a reduced GSH:GSSG ratio were the strongest discriminating factors between cases and controls. Measurement of total homocysteine, methionine, and total and reduced glutathione in reproductive aged women may play a role in primary prevention strategies targeted at CHDs.
本研究的目的是确定能够区分先天性心脏病（CHD）患儿母亲与未患先天性心脏病患儿母亲的代谢、遗传和生活方式因素。
我们分析了417例先天性心脏病患儿母亲和250例对照者的13种血清生物标志物浓度、3种功能性基因变异以及4种生活方式因素。为了确定区分病例组和对照组的危险因素，我们使用逻辑回归，随后进行递归划分以识别非线性相互作用。构建受试者工作特征（ROC）曲线以评估最终模型的鉴别准确性。
多种危险因素组合能够区分先天性心脏病患儿母亲与未患先天性心脏病患儿母亲。在21种可能的决定因素中，同型半胱氨酸和蛋氨酸的血清浓度以及还原型谷胱甘肽与氧化型谷胱甘肽的比值（GSH:GSSG）具有最大的鉴别能力。递归划分建模产生了5个终末节点，每个节点都说明了这三种生物标志物的相互作用。同型半胱氨酸水平升高且GSH:GSSG较低的女性，生育先天性心脏病患儿的风险最高，而同型半胱氨酸水平较低、蛋氨酸水平较高且GSH:GSSG较高的女性风险最低。ROC曲线下相应面积为81.6%（95%置信区间[CI]，78.1-85.2%），表明区分病例组和对照组的能力较高。
高同型半胱氨酸、低蛋氨酸以及较低的GSH:GSSG比值是病例组和对照组之间最强的鉴别因素。测量育龄妇女的总同型半胱氨酸、蛋氨酸以及总谷胱甘肽和还原型谷胱甘肽可能在针对先天性心脏病的一级预防策略中发挥作用。
