Quality control in the determination of cortisol in plasma/serum by using, on every sample, two different three-step separation methods including ultrafiltration, restricted-access high-performance liquid chromatography and reversed-phase high-performance liquid chromatography, and contrasting results to immunoassays

Tests of HPLC columns with restricted access, polymer covered alumina, polymer, and different ODS phases showed that base-acid compatible ODS columns gave the best peak shapes of cortisol, internal standard, as well as of plasma/serum (P/S) matrix components. Further trials with cortisol in P/S showed that three separation steps were essential in order to obtain chromatographic data which were superior to immunoassay data. Also, sufficient confidence in results required determination of each sample with two newly developed separation methods: (a) pre-separation with a restricted access column, concentration of the desired cut with a 20 mm base-acid compatible ODS column, and analysis with a 250 mm column filled with the same ODS; (b) pre-separation with an ultrafilter followed by the last two steps in (a). For detection UV was preferred over fluorescence. This twin multistep chromatography showed that immunoassays were very treacherous in that they produced a spectrum of results ranging from good to untenable without any warning whatever about functionality. The measurement of official controls, with reference values derived via gas chromatography-isotope dilution mass spectrometry, also demonstrated the superiority of the double HPLC method.