|
|
Background: The study aimed to assess the analytical performance of the Access Sensitive Estradiol (SNSE2) Assay on a DxI800 (Beckman Coulter, Brea, CA, USA) and compared it with a Cobas E 601 (Roche Diagnostics, Penzberg, Germany).
Methods: SNSE2 was assessed for imprecision, accuracy, limit of blank (LoB), limit of detection (LoD), limit of quantitation (LoQ), linearity, interference, and carryover. Two hundred and fourteen samples were run on both instruments. Bland-Altman plots, Passing-Bablok regression, and concordance correlation coefficient (CCC) graphs were used for comparisons.
Results: Access SNSE2 showed appropriate assay performance characteristics in terms of imprecision, LoB, LoD, LoQ, linearity, and interference. The Bland-Altman analysis of DxI 800 yielded negative bias from Cobas E 601 and the deviations for E2 ≤ 150 pmol/L, 150 - 500 pmol/L, and ≥ 500 pmol/L were found as 0.8%, -15%, and -8.9%, respectively. DxI 800 and E170 systems showed poor agreement for E2 levels ≤ 150 pmol/L and 150 - 500 pmol/L with CCC values of 0.7404 and 0.8342. For E2 levels ≤ 150 pmol/L there was a significant amount of both proportional and constant error with the highest slope of 1.518 (1.269 to 1.761) and an intercept of -45.08 (-66.09 to 18.78, respectively, according to the Passing-Bablok regression analysis).
Conclusions: Analytical performance for SNSE2 assay was found appropriate. However, attempts to improve harmonization and standardization across assays do not seem to contribute much for E2 measurements. Results obtained with different systems cannot be used interchangeably and follow up of patients should be done with the same system.
DOI: 10.7754/Clin.Lab.2020.200806
|
