Objective: To evaluate the analytical and clinical performance of an automated assay for serum cortisol on the ACS: 180 ®.
Methodology: Analytical performance (sensitivity, precision, linearity, carryover, interference) was assessed using standard laboratory protocols. The values obtained on the ACS for protocol specimens were compared to those obtained with a number of other automated assays on the ES300, TDx FL>., and Stratus.
Results:
The assay exhibited within-run and total coefficients of variation of <4% and 8.1%, respectively. Good correlation was observed among all the assays for patient specimens for a.m. and p.m. cortisol values; however, the mean values for the ACS were approximately 10-15% higher than those with the ES300 and Stratus and 5% lower than those with the TD x FLx. The reference range for all 4 assays as determined in a local population were higher than those claimed by the manufacturers. Good agreement among the assays were obtained for patients undergoing dexamethasone suppression and ACTH stimulation tests.
Conclusion:
The cortisol assay on the ACS:180 is accurate and precise and provides results that are clinically comparable to other automated assays. The random access capabilities of the instrument can facilitate workstation consolidation for improved laboratory efficacy.
times (2). Nonisotopic automated immunoassay procedures for cortisol employ enzyme, fluorescent, or chemiluminescent-mediated reactions. These assays generally exhibit improved sensitivity and specificity, as compared to the colorimetric assays, and are less time-consuming to perform than chromatographic methods. Most recently, a new automated chemiluminescent immunoassay for cortisol has become available on the ACS:180 (Ciba Corning Diagnostics, Walpole, MA, U.S.A.) (3-5). We evaluated the performance characteristics of this method and assessed its analytical and clinical performance in comparison to other commonly used automated immunoassays for cortisol. As an additional aid to clinical interpretation, reference intervals were evaluated with respect to time of collection.