Naslov
Usporedba dviju imunokemijskih metoda za određivanje troponina I na analizatoru Beckman Coulter UniCel DxI600
(Comparison of two imunochemical methods for troponin I determination on the Beckman Coulter UniCel DxI 600 analyser)

Autori
Jovanović, Marijana ; Starčić, Jelena ; Žarak, Marko ; Stančin, Nevenka

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, stručni

Izvornik
Biochemia Medica 2018 ; 28(Suppl 1):S1–S223 / - , 2018, 127-128

Skup
9. kongres hrvatskog društva za medicinsku biokemiju i laboratorijsku medicinu (HDMBLM)

Mjesto i datum
Zagreb, Hrvatska, 09.05.2018. - 12.05.2018

Vrsta sudjelovanja
Poster

Vrsta recenzije
Domaća recenzija

Ključne riječi
verifikacija, usporedba, troponin-I visoke osjetljivosti
(verification, comparison, high-sensitivity troponin-I)

Sažetak
Inroduction: High-sensitivity troponin I is a cardiac marker, used in early detection of myocardial infarction. The aim of the sudy is to compare the method for the high-sensitivity troponin I (Access hs TnI) with the daily routine method (Access AccuTnI+3 Beckman Coulter, CA, USA) used on the UniCel DxI 600 analyser, and to examine the coefficient of variation (CV) on the limit of quantification (LoQ) declared by the manufacturer. Materials and methods: A comparison was carried out on residual serum samples from 63 patients for whom troponin I testing was requested. Almost the entire range of the concentration range for Access hsTnI (5.6 – 27.027 ng/L) was examined. CV at LoQ was tested by repeated testing (N = 20) of the healthy respondents serum pool ; concentration 5.66 ng/L. Access hsTnI results in ng/L were converted to µg/L, because of data analysis. Due to abnormal distribution, the results were statistically analysed using Passing-Bablok regression analysis and Bland- Altman plot. Results: Passing-Bablok analysis showed regression line y = - 0.019 + 1.096 x (95% confidence interval (CI) for intersept A is - 0.02925 do - 0.01277, and for slope B 1.0419 do 1.249). Bland-Altman analysis showed an average BIAS of 4.2%, which meets the RiliBAK criteria (20%). All the samples outside the linearity of the method Access hsTnI were excluded from the statistical analysis. CV at LoQ was 15.8%, which doesn`t meet manufacturer criteria for CV (10%) at the concentration of 5.6 ng/L. Conclusion: The comparison results between the two methods indicate a constant and proportional error, but without clinical relevance. Although the results for trueness and precision meet the criteria (20%), for the full verification CV = 10% at the 99th percentile, reference intervals and linearity are yet to be confirmed. It can be assumed that CV at the 99th percentile will achieve 10% if CV on LoQ is 15.8%.

Izvorni jezik
Engleski

Znanstvena područja
Temeljne medicinske znanosti

POVEZANOST RADA