LC-MS/MS法与CLIA法检测伏立康唑血药浓度的结果比较

Comparison of serum voriconazole concentration by LC-MS/MS method and CLIA method

关键词:  伏立康唑  血药浓度  液相色谱-串联质谱法  化学发光免疫分析法

Key words:  voriconazole  blood concentration  liquid chromatography-tandem mass spectrometry  chemiluminescence immunoassay

摘要:

目的 评价液相色谱-串联质谱（liquid chromatography-tandem mass spectrometry，LC-MS/MS）法与化学发光免疫分析法（chemiluminescence immunoassay，CLIA）测定人血浆伏立康唑浓度结果的差异性、相关性和一致性，为临床监测伏立康唑血药浓度及个体化用药提供参考。方法 收集伏立康唑血药浓度达稳态后的患者血液样本153例，离心后取血浆，分别用LC-MS/MS和CLIA两种方法测定伏立康唑血药浓度，并对两种方法测定的结果进行统计学分析。结果 CLIA法测得的伏立康唑浓度4.61（2.95，6.67）μg/mL低于LC-MS/MS法测得的伏立康唑浓度4.85（3.44，6.66）μg/mL，且Wilcoxon配对检验结果显示差异有统计学意义（Z＝-4.175，P<0.001）；Spearman相关分析结果显示，CLIA法与LC-MS/MS法所测伏立康唑血药浓度的相关系数为0.970（P＜0.01）；Passing-Bablok回归分析结果显示，CLIA法与LC-MS/MS法检测伏立康唑血药浓度的回归方程为CCLIA＝-0.436+1.035CLC-MS/MS；Bland-Altman散点图分析结果显示，CLIA法与LC-MS/MS法所测的伏立康唑血药浓度的差值中有5.23％（8/153）的差值散点在一致性界限外（±1.96SD）；山形图的山顶对应的两种方法测得的差值也趋向于0；Kappa分析显示CLIA法和LC-MS/MS法检测伏立康唑血药浓度的临床符合率为94.12%（144/153），一致性系数Kappa=0.884。结论 CLIA法与LC-MS/MS法测定的伏立康唑血药浓度具有良好的相关性和一致性，但CLIA法测定的结果总体上低于LC-MS/MS法，因此临床用不同方法测定伏立康唑血药浓度时，对不同测定方法的差异应予以关注并作相应调整。

Abstract:

OBJECTIVE To evaluate the difference, correlation and consistency of the results of voriconazole concentration in human plasma determined by liquid chromatography-tandem mass spectrometry (LC-MS/MS) and chemiluminescence immunoassay (CLIA), so as to provide a reference for clinical monitoring of voriconazole blood concentration and individualized medication. METHODS Blood samples of 153 patients were collected after voriconazole blood concentration reached steady state, and plasma was obtained after centrifugation. The voriconazole blood concentration was determined by LC-MS/MS and CLIA, respectively, and the results of the two methods were statistically analyzed. RESULTS The voriconazole concentration measured by CLIA method was 4.61 (2.95, 6.67) μg/mL, which was lower than that by LC-MS/MS method, which was 4.85 (3.44, 6.66) μg/mL. And the Wilcoxon paired test showed that the difference was statistically significant (Z=-4.175, P<0.001). The Spearman correlation analysis showed that the correlation coefficient between the blood concentrations of voriconazole measured by CLIA and LC-MS/MS was 0.970 (P<0.01). The Passing-Bablok regression analysis showed that the regression equation of CLIA method and LC-MS/MS method for detection of voriconazole blood concentration was CCLIA=-0.436+1.035CLC-MS/MS. The Bland-Altman scatter plot analysis showed that 5.23% (8/153) of the difference of voriconazole concentration measured by CLIA method and LC-MS/MS method was outside the consistency limit (±1.96SD). The difference between the two methods corresponding to the top of the mountain plot also tended to 0. Kappa analysis showed that the clinical compliance rate of voriconazole blood concentration detected by CLIA method and LC-MS/MS method was 94.12% (144/153), and the consistency coefficient Kappa=0.884. CONCLUSION The blood concentrations of voriconazole determined by CLIA and LC-MS/MS showed good correlation and consistency, but the results of CLIA were generally lower than those of LC-MS/MS. Therefore, when different methods are used to determine the blood concentration of voriconazole in clinical practice, the differences between the different determination methods should be paid attention to and corresponding adjustments should be made.