A cross-sectional study of retrospective clinical liquid chromatography-tandem mass spectrometry detection consistency comparison program
10.3760/cma.j.cn114452-20250402-00211
- VernacularTitle:回顾性临床液相色谱串联质谱检测一致性比对计划横断面研究
- Author:
Wenda CHEN
1
;
Jiexin LIU
1
;
Yingfei PENG
1
;
Fangjun CHEN
1
;
Jiaqian QIN
1
;
Baishen PAN
1
;
Beili WANG
1
;
Wei GUO
1
Author Information
1. 复旦大学附属中山医院检验科，上海 200032
- Publication Type:Journal Article
- Keywords:
Liquid chromatography-tandem mass spectrometry;
Harmonization comparison;
Clinical testing;
External quality assessment
- From:
Chinese Journal of Laboratory Medicine
2025;48(12):1565-1570
- CountryChina
- Language:Chinese
-
Abstract:
Objective:? To summarize the achievements in improving the consistency of clinical liquid chromatography-tandem mass spectrometry (LC-MS/MS) testing results.Methods:? From 2021 to 2024, Zhongshan Hospital Affiliated to Fudan University recruited laboratories voluntarily participating in the MSHP (Clinical LC-MS/MS Testing Consistency Program). As of Batch 202404, a total of 76 laboratories had enrolled, including 60 medical institutions (all tertiary hospitals) and 16 third-party laboratories. Test items were established, and comparative samples were distributed regularly-each item′s samples covered three concentrations (high, medium, and low). Samples were shipped via cold chain and tested within one week. Our laboratory′s measurements served as the target, with participating labs′ results within ±25% of the target deemed qualified. Passing-Bablok regression and Bland-Altman analysis were used to assess consistency.Results:Taking 3-MT (3-methoxytyramine) as an example, the coefficients of variation (CVs) for the project′s three concentration levels improved from 17.00%, 47.18%, and 4.88% in the first comparative batch to 9.59%, 9.59%, and 6.1% in Batch 202404. Passing-Bablok regression results for the 5 units participating in 3-MT testing showed that Laboratory A had proportional bias but no systematic bias (regression slope [95% CI]: 0.903 [0.862-0.952]; intercept [95% CI]: 0.912 [-1.921-6.073]). The remaining laboratories exhibited no proportional or systematic bias with the target (Laboratory B: slope 1.031 [0.961-1.147], intercept-0.733 [-4.641-8.272]; Laboratory C: slope 0.982 [0.940-1.009], intercept-0.576 [-2.675-1.891]; Laboratory D: slope 0.973 [0.939-1.066], intercept-1.168 [-6.108-1.649]; Laboratory E: slope 0.999 [0.905-1.051], intercept-1.876 [-6.111-3.508]). Bland-Altman analysis indicated that all 5 laboratories′ results generally showed good consistency with the target. Through quality feedback and optimizing sample preparation concentrations, result consistency was enhanced.? Conclusion:? Clinical LC-MS/MS testing consistency programs contribute to improving the comparability of test results.
