First international descriptive and interventional survey for cholesterol and non-cholesterol sterol determination by gas- and liquid-chromatography–Urgent need for harmonisation of analytical methods

Abstract Serum concentrations of lathosterol, the plant sterols campesteroland sitosterol and the cholesterolmetabolite 5α- cholestanolare widely used as surrogate markers of cholesterolsynthesis and absorption, respectively. Increasing numbers of laboratories utilize a broad spectrum of well-established and recently developed methods for the determination of cholesteroland non- cholesterol sterols( NCS). In order to evaluate the quality of these measurements and to identify possible sources of analytical errors our group initiated the first international survey for cholesteroland NCS. The cholesteroland NCSsurvey was structured as a two-part survey which took place in the years 2013 and 2014. The first survey part was designed as descriptive, providing information about the variation of reported results from different laboratories. A set of two lyophilized pooled sera (A and B) was sent to twenty laboratories specialized in chromatographic lipid analysis. The different sterolswere quantified either by gas chromatography-flame ionization detection, gas chromatography- or liquid chromatography-mass selective detection. The participants were requested to determine cholesteroland NCSconcentrations in the provided samples as part of their normal laboratory routine. The second part was designed as interventional survey. Twenty-two laboratories agreed to participate and received again two different lyophilized pooled sera (C and D). In contrast to the first international survey, each participant received standard stock solutions with defined concentrations of cholesteroland NCS. The participants were requested to use diluted calibration solutions from the provided standard stock solutions for quantification of cholesteroland NCS. In both surveys, each laboratory used its own internal standard (5α- cholestane, epicoprostanol or deuterium labelled sterols). Main outcome of the survey was, that unacceptably high interlaboratory variations for cholesteroland NCSconcentrations are reported, even when the individual laboratories used the same calibration material. We discuss different sources of errors and recommend all laboratories analysing cholesteroland NCSto participate in regular quality control programs.