【摘要】 目的探讨强直性脊柱炎(AS)患者血清外源性代谢物与对照的差异,寻找AS的环境危险因素。方法用超高效液相色谱-质谱对49例AS患者及28名健康志愿者(HC)的外周血血清进行代谢组学检测。联合随机森林模型和单变量统计分析得出差异代谢物,分析其中外源性代谢物的特点及其与患者的CRP及ESR的相关性。结果各样本的检测过程稳定,保留时间变异度低。随机森林模型的总体预测错误率为0.143,处于较低水平,模型建立成功。结合单变量统计分析,得到10个校正后错误发现率<0.5的外源性代谢物,其中乙霉威、Avenanthramide a及二苯并呋喃的标准化峰强度两两之间均具有高度相关性(r均> 0.95,P均<0.01),二苯并呋喃、Avenanthramide a、BAXGP、乙霉威的标准化峰强度与CRP及ESR均呈正相关(P均<0.05)。结论血清代谢组学可在一定程度上反映AS的环境危险因素,外源性代谢物可能与AS患者的病情相关。更多还原

【Abstract】 Objective To identify the environmental risk factors of ankylosing spondylitis(AS) by comparing the serum exogenous metabolites between AS patients and healthy controls. Methods Ultra performance liquid chromatography-mass spectrometry(UPLC-MS) was applied to detect the serum metabolic profiles in 49 AS patients and 28 healthy controls. Random forest(RF) model combined with univariate statistical analysis were utilized to identify the differential metabolites between two groups. The characteristics of exogenous metabolites and their correlation with C reactive protein(CRP) and erythrocyte sedimentation rate(ESR) were studied. Results Lowretention time deviation of each sample suggested that the detection process was stable. The overall prediction error rate was 0.143, indicating that the RF model was successfully established. After the combination with univariate statistical analysis, 10 exogenous metabolites with an adjusted error detection rate < 0.5 were obtained. The normalized peak intensities of any two of diethofencarb, Avenanthramidea and dibenzofuran were significantly correlated(all r > 0.95,all P < 0.01). The normalized peak intensities of dibenzofuran, Avenanthramide a, BAXGP and diethofencarb were positively correlated with CRP and ESR(all P < 0.05). Conclusions Serum metabolomics can reflect the environmental risk factors of AS to certain extent. The exogenous metabolites might be associated with the severity of AS.更多还原