妊娠子痫前期患者血清代谢轮廓分析

Analysis of serum metabolic profiling of preeclampsia pregnancy

目的研究妊娠子痫前期患者血清代谢轮廓的变化，建立疾病区分模型并筛选出具有潜在临床诊断价值的特征代谢物。方法病例对照研究。收集2014年8月至2016年1月天津市第三中心医院产科住院确诊的31例子痫前期患者和25名正常妊娠分娩者及29名同期健康体检育龄女性，应用超高效液相色谱与质谱联用仪（UPLC－MS）分别对3组血清样本进行分析，采用基于模式识别的多元统计学方法对实验数据进行分析并对筛选出的差异性特征代谢物变化趋势进行分析。结果构建了妊娠子痫组血清代谢轮廓主成分分析模型及正交偏最小二乘判别模型，同时从模型中筛选并鉴定15种相比对照组在子痫患者血清中存在差异的特征代谢物，其中8种甘油磷脂类物质（包括7种溶血磷脂酰胆碱及1种溶血磷脂酸）及1种鞘磷脂类物质（植物鞘氨醇）在子痫前期组血清相对含量对比正常妊娠组之间有相同变化趋势，差异有统计学意义（Z值分别为2.32、3.34、3.21、2.60、2.22、3.40、3.58、5.84、2.70, P均〈0.05）。子痫前期组中1，25－二羟基－D3－26.23内酯和24－氧代－1－α，23，25二羟基维生素D3的相对含量明显低于正常妊娠组及对照组，差异具有统计学意义（Z值分别为2.01、3.89、3.26、2.34，P均〈0.05）。结论构建代谢轮廓判别模型具有很强的区分子痫前期患者与正常妊娠及健康育龄妇女的能力，筛选出特征代谢物能够早期反映子痫前期患者脂、钙磷代谢的状态，为子痫患者的预测、诊治等提供参考和帮助。（中华检验医学杂志，2017, 40：186－190）

Objectives This research explored the characteristics of changes in the serum metabolic profile of preeclampsia pregnancy（PE） to establish the disease distinguish model and screen characteristic metabolic markers with potential diagnostic value for preeclampsia. Methods From August 2014 to January 2016, samples in three groups were collected at Tianjin Third Central Hospital. Thirty-one clinically diagnosis patients with preeclampsia, 25 normal pregnancy women and 29 healthy volunteers of childbearing age were enrolled. Ultraperformance liquid chromatography-mass spectrometry （UPLC-MS） was used to analyze serum metabolites of PE group （31 patients with preeclampsia）, P group （25 normal pregnancy women） and Normal group （29 healthy volunteers of childbearing age） . Nonparametric test analyzes were used to analyze the data and find the specific metabolites. Results This research established the principal component analysis （PCA） and orthogonal partial least squares-discriminant analysis （OPLS-DA） disease distinguish model for PE group, P group and Normal group. To distinguish PE group, P group and Normal group, 15 characteristic metabolites were identified. Eight kinds of glycerol phospholipid （ including 7 kinds of hemolysis phosphatidyl choline and 1 kind of lysophospholipids acid） and 1 kind of sphingomyelin in PE group were higher than that of normal pregnancy group. The difference had statistically significant（Z of the metabolites were 2. 32, 3.34, 3.21,2. 60, 2. 22, 3.40, 3.58, 5.84, 2.70 respectively,all P 〈 0.05）. 1, 25-Dihydroxyvitamin D3-26,23-1actone and 24-Oxo-lalpha, 23,25-trihydroxyvitamin D3 in PE group were higher than that of P group and Normal group, which had a statistics difference （ Z of the metabolites were 2.01, 3.89, 3.26, 2. 34 respectively, all P 〈 0. 05 ）. Conclusions Metabolomics distinguish model has a good ability to distinguish PE group, P group and Normal group. Serum characteristic metabolites can successfully reflect the status of fat, calcium and phosphorus metabolism of preeclampsia patients and provide high value for prediction, diagnosis and treatment. （ Chin J Lab Med, 2017, ,40： 186-190）