地理环境因子对肌酸激酶同工酶参考值的影响  

Effect of geographical factors on reference values of creatine kinase isoenzyme

作　　者：庞欣蕊 葛淼 王晶 张雷 李佳鑫 王聪霞[2] 何进伟[3] PANG Xinrui;GE Miao;WANG Jing;ZHANG Lei;LI Jiaxin;WANG Congxia;HE Jinwei(Institute of Health Geography,School of Geography and Tourism,Shaanxi Normal University,Xi'an 710119,China;Second Affiliated Hospital of Xi'an Jiaotong University,Xi'an 710004,China;Medical College of Yan'an University,Xi'an 716099,China)

机构地区：[1]陕西师范大学地理科学与旅游学院健康地理研究所,西安710119 [2]西安交通大学医学部第二附属医院,西安710004 [3]延安大学医学院,西安716099

出　　处：《地理研究》2022年第6期1700-1714,共15页Geographical Research

基　　金：国家自然科学基金项目(41761100);陕西省自然科学基础研究计划项目(2019JM-408)。

摘　　要：研究目的旨在探讨三种常用的变量筛选回归方法对中国健康成年人肌酸激酶同工酶(CK-MB)参考值在地理空间上的分布规律特征,应用于制定不同地区CK-MB参考值标准,为临床医学研究做出贡献。方法通过阅读大量文献搜集全国137个市县级单位共8697例健康成年人CK-MB参考值,并选择了地理位置、气候、土壤三大类共24项地理因子。使用相关分析方法检验CK-MB参考值和地理因子之间的显著性,提取到9项相关性地理因子。基于R语言评估模型多重共线性的严重程度,建立CK-MB参考值岭回归(Ridge)模型、拉索回归(Lasso)模型、主成分分析模型。对比得到最优预测模型,拟合全国2322个市县级单位健康成年人CKMB预测参考值,再结合地统计分析,运用析取克里金法进行趋势分析,得到CK-MB预测参考值地理分布规律特征。结果表明全国2322个市县级单位的健康成年人CK-MB参考值具有空间自相关性,模型测试表明,主成分分析法具有更好的模拟和预测能力。研究表明健康成年人CK-MB参考值分别与纬度、年日照时数、年平均相对湿度、年降水量、气温年较差、年平均气温、表土石砾含量、表土(粘土)阳离子交换量、表土(粉土)阳离子交换量具有显著的相关性。由地理空间分布图显示,整体上呈北高南低,从东南沿海向西北内陆逐渐升高的趋势。通过搜集到的任一地区地理因子,结合主成分分析预测模型或已得的地理分布图可确定该地区健康人CK-MB参考值范围,为在临床诊断中考虑地域差异提供参考。This paper aims to explore the geographical distribution characteristics of the reference values of creatine kinase isoenzyme in healthy Chinese adults by three commonly used variable-screening regression methods,and to apply them to the formulation of the reference values of CK-MB,so as to make contributions to clinical medicine research.The reference values of CK-MB for 8697 healthy adults in 137 cities and county-level units were collected according to a large number of literates,and 24 geographical factors covering terrain,climate and soil were selected.Correlation analysis method was used to test the significance between CK-MB reference value and geographical factors,and 9 geographical factors were extracted.Based on R language to evaluate the degree of multi col-linearity of the model,CKMB Ridge model,Lasso model and principal component analysis model were established.The optimal prediction model was obtained by comparison,and the reference values of CK-MB for healthy adults in 2322 city and county units in China were fitted.Then the combined statistical analysis was conducted,and the disjunctive Kriging method was used for trend analysis to explore the geographical distribution characteristics of the reference values of CK-MB.The CK-MB reference values of healthy adults in 2322 city and county units in China are characterized by the pattern of"higher in the north and lower in the south",and gradually increases from the southeast coastal area to the northwest inland area.The reference value of CK-MB for healthy people in the region can be determined by the collected geographical indicators of any region,combined with the principal component analysis prediction model or the obtained geographical distribution map,so as to provide a reference for studying regional differences in clinical diagnosis.

关 键 词：肌酸激酶同工酶(CK-MB) 空间自相关 岭回归(Ridge)模型 拉索回归(Lasso)模型 主成分分析 地统计分析 

分 类 号：R446.1[医药卫生—诊断学]