engleski

Determination of free cortisol before and after extraction in 24-h urine samples

BACKGROUND-AIM: Measuring urine free cortisol level is an important step in Cushing’s syndrome diagnosis. In our laboratory, routineanalysis of cortisol in 24-hour urine is performed on an automated immunoassay Beckman Coulter DxI 600 analyzer (CLIA), preceded by cortisol extraction with ethyl acetate. With this step, water-soluble metabolites are removed. On the other hand, this procedure is time-consuming and prone to preanalytical errors due to manual work. Patientsamples are collected, stored frozen, and analyzed once a week. Furthermore, ethyl acetate is an irritant and aflammable chemical with a special disposal process. The aim pf this summary was to examine whether analyzing urinedirectly without extraction could be an alternative. METHODS: For that purpose 26 patient samples were analyzed immediately after the reception, with and without extraction. Inaddition, urine cortisol without extraction was measured for 52 patient samples and compared with the results withoutextraction obtained on Abbott Architect i1000SR (CMIA) whose calibrators are traceable to the LC-MS/ MS method, andthe LC-MS/MS calibration is verified by certified reference materials. RESULTS: A Passing-Bablok regression analysis of results with and without extraction reveled a regression equation: y =-257.60+ 4.61 x (95% CI for intercept: -1356.58 to 27.32 ; 95% CI for slope: 2.27 to 13.57), from which is obvious thatthe methods are not comparable. These differences were expected and further comparisons were made.The obtained Passing-Bablock equation for CMIA-CLIA comparison was: y= 87.34+3.22x (95% CI for intercept: 38.31 to134.23 ; 95% CI for slope: 2.73 to 3.63), which indicates a statistically significant constant and proportional deviation inthe CMIA (Abbott) and CLIA (Beckman Coulter) methods. Given the differences between the reference ranges of thesemethods, the results can only be compared at the clinical decision level. In this context, the obtained results fulfilledthe acceptance criteria for the kappa coefficient, with a percentage of agreement of 100% (kappa = 1.0). CONCLUSIONS: In order to reduce toxic laboratory waste and the danger of poisoning or fire, we decided from now on to determinefree urinary cortisol without prior extraction on Abbott Architect i1000SR due its traceability to LC-MS/MS.

free cortisol ; 24-h urine samples ; extraction ; immunochemistry

nije evidentirano