激光诱导击穿光谱定量分析全血中的锂元素  

Quantitative Analysis of Lithium Element in Whole Blood Using Laser-induced Breakdown Spectroscopy

作　　者：任文心 杨亮 赵韩 王奕朦 黄达 戴鑫华 林庆宇[1] 段忆翔[1] REN Wen-Xin;YANG Liang;ZHAO Han;WANG Yi-Meng;HUANG Da;DAI Xin-Hua;LIN Qing-Yu;DUAN Yi-Xiang(School of Mechanical Engineering,Sichuan University,Chengdu 610065,China;Chengdu Aliben Science&Technology Co.,Ltd.,Chengdu 611930,China;West China School of Medicine,Sichuan University,Chengdu 610041,China;Department of Experimental Medicine,West China Hospital of Sichuan University,Chengdu 610041,China)

机构地区：[1]四川大学机械工程学院,成都610065 [2]成都艾立本科技有限公司,成都611930 [3]四川大学华西临床医学院,成都610041 [4]四川大学华西医院实验医学科,成都610041

出　　处：《分析化学》2024年第4期559-565,共7页Chinese Journal of Analytical Chemistry

基　　金：四川省科技厅重点研发项目(No.2023YFSY0044)资助。

摘　　要：锂(Li)盐作为一种精神科药物常被用于治疗重性抑郁障碍,然而长期服用Li盐会产生高中毒风险,及时监测患者血液中Li元素的浓度,对确保用药安全和临床治疗效果非常重要。激光诱导击穿光谱(Laser-induced breakdown spectroscopy,LIBS)作为一种快速分析技术,被广泛应用于实际场景下复杂基质的元素分析。本研究采用LIBS技术结合偏最小二乘法(Partial least squares,PLS)对血液基质中的Li元素进行定量分析。采用45组临床血液样品,分别构建血浆和全血基质下Li元素的定量分析模型,通过五折交叉验证方法对PLS算法中的潜变量数进行优化。结果表明,基于血浆基质构建的PLS定量分析模型预测决定系数(R2)为0.992,预测均方根误差(RMSEP)为0.204μg/mL,相对标准偏差(RSD)为2.14%;基于全血基质构建的PLS定量分析模型R2为0.984,RMSEP为0.728μg/mL,RSD为3.45%,因此,基于血浆定标值所建立的LIBS模型定量分析全血中锂元素的效果更佳。LIBS技术为临床血锂的快速检测评估提供了一种新的选择,具有良好的应用前景。Lithium(Li) salts are commonly used as psychotropic medications for the treatment of major depressive disorders.However,long-term use of Li salts poses a high risk of toxicity,necessitating continuous monitoring of Li concentration in patient blood to ensure medication safety,which is crucial for clinical treatment.Laser-induced breakdown spectroscopy(LIBS),as a rapid analytical technique,has been widely applied in the elemental analysis of complex matrices in various practical scenarios.In this study,LIBS technology combined with partial least squares(PLS) was employed for quantitative analysis of Li elements in blood matrix.A total of 45 clinical blood samples were utilized,and the quantitative models for plasma and whole blood matrices were separately investigated.The number of latent variables in the PLS algorithm was optimized using a five-fold cross-validation method.Results revealed that the PLS quantitative model constructed on the basis of plasma matrix achieved a predictive determination coefficient(R2) of 0.992,a predictive root mean square error(RMSEP) of 0.204μg/mL,and a relative standard error(RSD) of 2.14%.In contrast,for the PLS quantitative model constructed on the basis of whole blood matrix,theR2was 0.984,the RMSEP was 0.728μg/mL,and the RSD was 3.45% Consequently,the LIBS model constructed on the basis of plasma calibration values demonstrated superior performance in quantitative analysis of Li element in whole blood,and LIBS technology provided a new possibility for rapid assessment of blood Li levels in clinical practice,with promising prospects for application.

关 键 词：激光诱导击穿光谱 血液 锂元素 定量分析 偏最小二乘法 

分 类 号：O657.319[理学—分析化学] R969.1[理学—化学]