Analytical Performance of the Fully Automated AxSYM(r) Estradiol Assay

Abstract

The test performance of the fully automated AxSYM^® Estradiol available on the immunoassay analyser Abbott AxSYM^® was evaluated. Imprecision, sensitivity and linearity of dilution were examined. For assessment of accuracy, the assay results measured in 12 control pools and 9 patient samples were compared with the values obtained with isotope dilution—mass spectrometry (reference method). The correlation with the manual radioimmunoassay Estradiol MAIA^® was evaluated using 140 serum samples with estradiol concentrations ranging from 0 to 10 nmol/l. Imprecision studies revealed for the AxSYM^® Estradiol within-run coefficients of variation of 2.8–8.4% and day-to-day coefficients of variation of 3.4–9.5% (concentration range 0.3–2.8 nmol/l). The lower limit of quantification (lowest estradiol concentration with a day-to-day coefficient of variation <20%) was <0.11 nmol/l; the lower limit of detection was <0.05 nmol/l. The estradiol concentration recovered in patient samples after dilution did not differ by more than 10% from the expected values. The estradiol concentration measured with the AxSYM^® Estradiol in 12 commercial control pools in some cases grossly deviated from the reference method values; however, for 9 individual patient samples the AxSYM results deviated by not more than 22% (−11%−+22%). There was a good overall correlation (coefficient of correlation = 0.989) between the results measured with the AxSYM^® Estradiol and the Estradiol MAIA^® in patient samples.

The AxSYM^® Estradiol assay exhibits a good precision and a sufficient degree of sensitivity for measurement of estradiol in serum of menstruating women. Although the AxSYM results measured in the control pools, in some cases did not meet the target values, the good correlation with the Estradiol MAIA^® indicates that the reliability of the AxSYM^® Estradiol for clinical practice is comparable with well established radioimmunoassays. Thus, the AxSYM^® Estradiol offers an alternative which is comparable with respect to clinical reliability but has great advantages in view of rapidity, flexibility and convenience of analysis.