Improvements in the simultaneous determination of calcidiol and calcitriol in human serum or plasma.

Technical improvements have been applied to reduce the time required for the determination of calcidiol and calcitriol in serum or plasma. The modifications include the use of Sep-Pak C18 cartridges for the extraction of calciol metabolites from serum instead of a classical liquid/liquid extraction, a considerable shortening of the HPLC purification time compared with our previously described method, and the application of HPLC with UV detection at 254 nm of calcidiol as an alternative to the usual competitive protein binding methods. In none of the 199 samples where calcidiol was determined by HPLC did we observe a detectable peak of ercalcidiol. Quantitation of calcidiol by HPLC and competitive protein binding was compared in 5 series of assays. The correlation between the two methods was 0.99. The average slope of the linear regression line when the HPLC values were plotted versus the competitive protein binding values was 1.14. The average intercept was 0.19 nmol/l. The mean within-run coefficient of variation for calcidiol in these series was 5% for the competitive protein binding method, and 4% for HPLC method. Between-run coefficients of variation were 6% and 12% for the competitive protein binding and for the HPLC method, respectively. Within-run and between-run coefficients of variation for calcitriol were 6% and 15%, respectively.